I
n
t
ern
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
E
l
ect
ri
ca
l
a
n
d
C
o
m
p
u
t
er E
n
g
i
n
eeri
n
g
(
I
J
E
C
E
)
V
o
l.
11
,
N
o.
6
,
D
ecem
b
er
202
1
,
pp.
54
20
~
5
429
I
S
S
N
:
2088
-
8708
,
D
O
I
:
10.
11
591/
i
j
ece
.
v1
1
i
6
.
pp
54
20
-
542
9
5420
Jou
r
n
al
h
om
e
p
age
:
h
ttp
:
//ije
c
e
.
ia
e
s
c
o
r
e
.
c
o
m
I
m
a
g
e m
ul
t
i
-
lev
el
-
t
hres
ho
lding
w
it
h M
a
y
f
ly
o
pt
i
m
i
za
t
io
n
Se
i
f
e
di
ne
K
a
dr
y
1
,
V
e
n
k
a
t
e
s
a
n R
a
j
i
ni
k
a
nt
h
2
,
Ja
m
i
n
K
oo
3
,
B
ye
on
g
-
G
w
o
n K
a
ng
4
1
D
ep
ar
t
m
en
t
o
f
M
at
h
em
at
i
cs
an
d
C
o
m
p
u
t
er
S
ci
en
ce,
F
acu
l
t
y
o
f
S
ci
en
ce,
B
ei
r
u
t
A
r
ab
U
n
i
v
er
s
i
t
y
,
L
eb
an
o
n
2
D
e
pa
r
t
m
e
nt
of
E
l
e
c
t
r
oni
c
s
a
n
d I
ns
t
r
um
e
nt
a
t
i
on E
ng
i
ne
e
r
i
ng
,
S
t
.
J
os
e
ph’
s
C
ol
l
e
g
e
of
E
ng
i
ne
e
r
i
ng
,
C
he
nna
i
,
I
n
di
a
3
,4
D
ep
ar
t
m
en
t
o
f
I
nf
or
m
a
t
i
on
a
nd C
om
m
uni
c
a
t
i
on T
e
c
hnol
og
y
(
IC
T
)
C
onv
e
r
g
e
nc
e
,
S
oonc
h
un
hy
a
ng U
ni
v
e
r
s
i
t
y
,
A
s
a
n,
S
out
h K
or
e
a
A
rt
i
cl
e I
n
f
o
AB
S
T
RAC
T
A
r
tic
le
h
is
to
r
y
:
R
ecei
v
ed
De
c
4
, 2
0
20
Re
v
i
se
d
Ap
r
5
,
20
21
A
ccep
t
ed
Ap
r
26
,
20
21
I
m
a
g
e
t
hr
e
s
hol
di
ng
i
s
a
w
e
l
l
a
ppr
ov
e
d pr
e
-
pr
oc
e
s
s
i
ng
m
e
t
hod
ol
og
y
a
nd
e
nha
nc
i
ng
t
he
i
m
a
g
e
i
nf
or
m
a
t
i
o
n ba
s
e
d o
n a
c
hos
e
n t
hr
e
s
ho
l
d
i
s
a
l
w
a
y
s
p
r
ef
er
r
ed
.
T
h
i
s
r
es
ear
ch
i
m
p
l
e
m
en
t
s
t
h
e
m
a
y
f
l
y
o
p
tim
iz
a
tio
n
a
lg
o
r
ith
m
(
M
O
A
)
b
as
ed
i
m
ag
e
m
u
lti
-
l
ev
el
-
t
hr
e
s
hol
di
ng
on a
cl
as
s
o
f
b
en
ch
m
ar
k
i
m
a
g
e
s
of
di
m
e
ns
i
on 5
12x
5
12x
1
.
T
he
M
O
A
i
s
a
nov
e
l
m
e
t
hod
ol
og
y
w
i
t
h t
he
a
l
g
or
i
t
hm
pha
s
e
s
,
s
uc
h a
s
;
i
)
I
n
iti
a
liz
a
tio
n
,
ii
)
E
x
pl
or
a
t
i
on
w
i
t
h
ma
l
e
-
m
a
y
fl
y
(M
M
),
ii
i
)
Ex
p
lo
r
a
ti
o
n
w
ith
f
e
m
al
e
-
m
a
y
f
l
y
(F
M
),
iv
)
O
f
f
s
pr
i
ng
g
e
ne
r
a
t
i
on
an
d
,
v
)
T
e
r
mi
n
a
tio
n
.
T
h
is
a
lg
o
r
ith
m
i
m
p
le
m
e
n
ts
a
s
tr
ic
t t
w
o
-
s
t
ep
s
ear
ch
pr
oc
e
dur
e
,
i
n w
hi
c
h e
ve
r
y
M
a
y
f
l
y
i
s
f
or
c
e
d t
o a
t
t
a
i
n t
he
g
l
oba
l
be
s
t
s
ol
ut
i
o
n.
T
he
pr
opos
e
d r
e
s
e
a
r
c
h c
ons
i
de
r
s
t
he
t
hr
e
s
h
ol
d v
a
l
ue
f
r
om
2 t
o
5 a
nd
t
he
s
upe
r
i
or
i
t
y
o
f
t
he
r
e
s
ul
t
i
s
c
onf
i
r
m
e
d b
y
c
o
m
p
u
tin
g
th
e
e
s
s
e
n
tia
l I
m
a
g
e
q
u
al
i
t
y
m
ea
s
u
r
es
(
I
Q
M
)
.
T
h
e p
e
r
f
o
r
m
an
ce o
f
M
O
A
i
s
al
s
o
co
m
p
ar
ed
an
d
v
a
l
i
da
t
e
d a
g
a
i
ns
t
t
he
ot
he
r
pr
oc
e
dur
e
s
,
s
uc
h a
s
p
a
r
tic
le
-
s
wa
r
m
-
o
p
tim
iz
a
tio
n
(
PSO
)
,
b
a
c
te
r
ia
l f
o
r
a
g
in
g
o
p
tim
iz
a
tio
n
(
BF
O
)
,
fi
r
e
fl
y
-
a
l
g
o
r
ith
m
(
FA
)
,
b
at
a
lg
o
r
ith
m
(
BA
)
,
cu
c
k
o
o
s
ear
ch
(
CS
)
a
nd
mo
t
h
-
f
la
m
e
o
p
ti
m
iz
a
ti
o
n
(
MFO
)
a
nd t
he
a
t
t
a
i
ne
d p
-
v
a
l
ue
of
W
i
l
c
ox
on r
a
nk
t
e
s
t
c
onf
i
r
m
e
d t
he
s
upe
r
i
or
i
t
y
of
t
he
M
O
A
c
om
pa
r
e
d
w
i
t
h ot
he
r
a
l
g
or
i
t
hm
s
c
ons
i
de
r
e
d i
n t
h
i
s
w
or
k
.
Ke
y
wo
rd
s
:
F
eat
u
r
e
-
s
i
m
ila
r
it
y
-
i
nd
e
x
W
ilc
o
x
o
n
te
s
t
M
a
y
f
l
y
o
p
ti
m
iz
a
tio
n
O
t
su
T
hr
e
s
ho
l
d
i
ng
T
hi
s
i
s
an
ope
n ac
c
e
s
s
ar
t
i
c
l
e
u
nd
e
r
t
he
CC B
Y
-
SA
l
i
cen
s
e.
Co
rre
sp
o
n
d
i
n
g
Au
t
h
o
r
:
B
ye
o
n
g
-
G
wo
n
K
a
n
g
D
e
p
a
r
t
m
e
nt
o
f
I
C
T
C
o
nve
r
ge
nc
e
S
o
o
nc
h
un
h
ya
n
g U
ni
ve
r
s
i
t
y
A
s
a
n
31538,
S
ou
t
h
K
or
e
a
E
m
a
i
l
:
b
g
ka
ng
@
s
c
h.
a
c
.
kr
1.
I
NT
RO
D
UCT
I
O
N
R
ecen
t
l
y
,
a co
n
s
i
d
er
ab
l
e n
u
m
b
er
o
f
r
es
ear
ch
w
o
r
k
s
ar
e p
r
o
p
o
s
ed
an
d
i
m
p
l
e
m
e
n
t
ed
b
y
t
h
e
r
es
ear
ch
es
i
n
v
a
r
i
ous
do
m
a
i
n
s
,
i
n
w
hi
c
h
t
h
e
i
m
a
g
e
s
r
e
c
or
de
d
u
s
i
ng
a
c
h
os
e
n
pr
oc
e
du
r
e
c
om
m
onl
y
;
h
e
l
p t
o
a
s
se
s t
h
e
e
s
s
e
n
tia
l in
f
o
r
m
a
t
io
n
[
1
]
-
[
5
]
.
I
m
ag
e p
r
o
ces
s
i
n
g
i
s
e
m
er
g
ed
as
o
n
e o
f
t
h
e k
e
y
r
es
ear
ch
dom
a
i
ns
a
nd
w
i
d
e
l
y
a
dopt
e
d
t
o
pr
oc
e
s
s
r
e
d
,
gr
e
e
n,
d
a
n
b
l
ue
(
RG
B
)
/
gr
a
ys
c
a
l
e
i
m
a
ge
s
w
i
t
h
c
ho
s
e
n
m
e
t
ho
d
o
l
o
gi
e
s
[
6
]
-
[
8
].
E
v
e
n
t
h
ough
a
c
ons
i
de
r
a
bl
e
num
b
e
r
of
i
m
a
g
e
pr
oc
e
s
s
i
n
g a
ppr
oa
c
h
e
s
e
x
i
s
t
,
t
h
e
im
a
g
e
m
u
lti
-
l
e
ve
l
t
hr
e
s
ho
l
d
i
n
g
(
I
M
L
T
)
i
s
e
m
er
g
ed
as
o
n
e o
f
t
h
e
v
i
t
al
m
et
h
o
d
s
t
o
p
r
o
ces
s
v
ar
i
o
u
s
c
l
as
s
i
m
ag
e
s
w
i
t
h
as
s
i
g
n
ed
t
h
r
es
h
o
l
d
va
l
ue
s
[
9
]
-
[
1
2
]
.
I
n
m
o
s
t
o
f
t
h
e
cas
es
,
t
h
e
I
M
L
T
i
s
co
n
s
i
d
er
e
d
t
o
en
h
a
n
ce t
h
e i
m
ag
e
t
ex
t
u
r
e/
p
i
x
el
i
n
f
o
r
m
at
i
o
n
by
g
r
ou
pi
n
g
a
l
i
k
e
pi
x
e
l
s
ba
s
e
d on
t
h
e
c
h
os
e
n
t
h
r
e
s
h
ol
ds
.
I
t
c
a
n
be
i
m
pl
e
m
e
nt
e
d
w
i
t
h
r
e
c
o
m
m
e
n
de
d
m
e
t
h
ods
,
s
uc
h a
s
O
t
s
u,
K
a
p
ur
,
S
ha
nno
n
a
nd
T
s
a
l
l
i
s
w
i
t
h c
o
m
m
o
n va
l
ue
s
o
f
p
r
e
f
e
r
r
e
d
t
hr
e
s
ho
l
d
s
(
T
) a
s
;
2
-
5 [
13
]
-
[
18]
.
F
r
o
m
y
ear
1
9
7
9
,
O
t
s
u
i
s
co
m
m
o
n
l
y
e
m
p
l
o
y
ed
t
o
e
n
h
a
n
ce t
h
e R
G
B
/
g
r
e
y
s
cal
e
i
m
a
g
es
b
as
ed
o
n
t
h
e
c
h
os
e
n
t
h
r
e
s
h
ol
d [
19]
.
D
u
e
t
o i
t
s
s
u
pe
r
i
or
i
t
y
,
t
h
i
s
a
ppr
oa
c
h
i
s
w
i
de
l
y
a
dopt
e
d t
o pr
e
-
p
r
o
ces
s
a cl
as
s
o
f
m
ed
i
cal
i
m
ag
e
s
r
eco
r
d
ed
w
i
t
h
d
i
f
f
er
e
n
t
m
o
d
a
litie
s
[
2
0
]
,
[
2
1
]
.
T
h
e ear
l
i
er
w
o
r
k
s
i
n
l
i
t
er
at
u
r
e al
s
o
co
n
f
i
r
m
t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
I
S
S
N
:
2088
-
8708
I
m
age
m
ul
t
i
-
l
evel
-
th
r
e
s
h
o
ld
in
g
w
ith
M
a
y
fly
o
p
tim
iz
a
tio
n
(
Se
i
f
e
di
ne
K
adr
y
)
5421
ne
e
d
o
f
O
t
s
u’
s
t
hr
e
s
ho
l
d
i
ng
f
o
r
m
e
d
i
c
a
l
i
m
a
ge
e
xa
m
i
na
t
i
o
n
u
s
i
n
g
t
he
m
a
c
hi
ne
-
l
ear
n
i
n
g
an
d
d
eep
-
le
a
r
n
in
g
t
ech
n
i
q
u
es
[
2
2
]
,
[
23]
.
T
he
O
t
s
u
’
s
s
c
he
m
e
i
s
c
o
m
m
o
nl
y u
s
e
d
t
o
e
nha
nc
e
t
he
t
e
s
t
i
m
ag
e
s
b
as
ed
o
n
t
h
e c
h
o
s
en
t
h
r
es
h
o
l
d
an
d
it c
a
n
b
e
i
m
p
le
m
e
n
te
d
w
it
h
; i
)
T
r
a
d
itio
n
a
l a
n
d
ii
)
H
e
u
r
i
s
tic
a
l
g
o
r
ith
m
a
s
s
i
s
te
d
te
c
h
n
i
q
u
e
s
.
I
n
tr
a
d
itio
n
a
l
m
e
t
h
od,
a
m
a
nu
a
l
ope
r
a
t
or
i
s
a
l
l
o
w
e
d t
o pe
r
f
or
m
a
num
be
r
of
t
r
i
a
l
s
t
o v
a
r
y
t
h
e
t
h
r
e
s
h
ol
d of
t
h
e
i
m
a
g
e
t
i
l
l
t
he
b
et
w
ee
n
-
c
l
a
s
s
-
v
ar
i
an
ce
(B
C
V
) i
s
i
m
p
ro
v
e
d
.
In
h
e
u
ri
s
t
i
c
-
a
l
g
o
r
ith
m
m
et
h
o
d
,
a ch
o
s
e
n
co
m
p
u
t
er
al
g
o
r
i
t
h
m
i
s
p
e
r
m
i
tte
d
to
a
d
j
u
s
t t
h
e
i
m
a
g
e
t
h
r
e
s
h
o
ld
,
till i
m
p
r
o
v
e
d
v
a
lu
e
o
f
B
C
V
is
r
e
a
c
h
e
d
.
I
m
p
le
m
e
n
ta
tio
n
o
f
tr
a
d
itio
n
a
l
w
o
r
k
s
ee
m
s
t
o
b
e m
o
r
e co
m
p
l
ex
f
o
r
I
M
L
T
an
d
h
en
ce,
a
cl
as
s
o
f
h
eu
r
i
s
t
i
c
a
lg
o
r
it
h
m
-
b
as
ed
m
et
h
o
d
s
ar
e
w
id
e
l
y
a
d
o
p
te
d
in
r
e
c
e
n
t d
a
y
s
th
a
n
tr
a
d
itio
n
a
l th
r
e
s
h
o
ld
i
n
g
[
2
4
]
-
[
27]
.
T
h
e
pr
opos
e
d
w
or
k
a
i
m
s
t
o de
m
ons
t
r
a
t
e
t
h
e
I
M
L
T
o
pe
r
a
t
i
o
n
u
s
i
ng
t
h
e
r
e
c
e
nt
l
y
de
v
e
l
ope
d a
pp
r
oa
c
h
n
a
me
d
m
a
y
f
l
y
o
p
ti
m
iz
a
tio
n
a
lg
o
r
ith
m
(
M
O
A
)
[
28]
.
T
h
e
M
O
A
i
s
pr
opos
e
d i
n
2020 b
y
c
om
bi
n
i
n
g t
h
e
be
s
t
s
tr
a
te
g
ie
s
i
n
fi
r
e
fl
y
-
a
l
g
o
r
ith
m
(F
A
),
p
a
r
tic
le
-
s
wa
r
m
-
o
p
ti
m
i
z
a
tio
n
(
P
SO
)
,
a
nd
ge
ne
t
i
c
-
a
l
g
o
r
ith
m
(G
A
).
T
h
e
M
O
A
co
n
s
i
d
er
s
a
m
u
l
t
i
p
l
e s
ea
r
ch
s
t
r
at
eg
y
f
o
r
t
h
e o
p
t
i
m
al
s
o
lu
tio
n
u
s
i
n
g
th
e
a
r
ti
f
ic
ia
l M
a
y
f
lie
s
w
i
th
m
a
le
a
n
d
f
e
m
al
e cat
e
g
o
r
y
.
F
u
r
t
h
er
,
t
h
i
s
al
g
o
r
i
t
h
m
al
s
o
g
e
n
er
at
ed
a co
u
p
l
e o
f
M
a
y
f
l
i
es
(
m
al
e
a
n
d
f
e
m
al
e)
w
i
t
h
an
as
s
i
g
n
ed
v
el
o
ci
t
y
v
a
l
u
e o
f
zer
o
.
D
u
e t
o
t
h
e m
u
l
t
i
p
l
e s
ear
ch
p
r
act
i
ce ex
i
s
t
i
n
g
i
n
M
O
A
,
g
et
t
i
n
g
t
h
e f
i
n
e
st
r
e
su
l
t
b
as
ed
o
n
t
h
e as
s
i
g
n
ed
t
h
r
es
h
o
l
d
b
eco
m
es
a
n
eas
y
t
a
s
k
[
2
9
]
.
I
n t
hi
s
w
o
r
k,
t
he
I
M
L
T
i
s
i
m
p
l
e
m
e
nt
e
d
o
n
we
l
l
-
k
n
o
wn
g
r
e
y
s
c
a
l
e
i
m
a
g
e
s
w
i
t
h
di
m
e
ns
i
on
512x
512x
1
p
i
x
el
s
a
n
d
t
h
e a
t
t
ai
n
ed
r
es
u
l
t
s
ar
e t
h
e
n
co
m
p
ar
ed
ag
ai
n
s
t
t
h
e o
r
i
g
i
n
al
t
e
s
t
i
m
a
g
e t
o
co
m
p
u
t
e
th
e
e
s
s
e
n
tia
l
i
ma
g
e
-
q
u
a
lit
y
-
p
ar
a
m
et
er
s
(IQ
P
) [3
0
]
-
[
32]
.
T
h
e
pr
op
os
e
d w
or
k i
s
i
m
pl
e
m
e
n
t
e
d
w
i
t
h
T
=
2 t
o 5 a
n
d t
h
e
o
b
t
ai
n
ed
r
es
u
l
t
s
as
w
el
l
a
s
I
Q
P
s
ar
e t
ab
u
l
at
ed
.
I
n
o
r
d
er
t
o
m
i
n
i
m
i
ze t
h
e s
ear
c
h
t
i
m
e,
a
bou
n
de
d
-
t
hr
e
s
ho
l
d
-
s
ear
ch
(
B
T
S
)
is
e
m
p
lo
y
e
d
,
in
w
h
ic
h
t
h
e
th
r
e
s
h
o
ld
v
a
lu
e
s
w
i
th
t
h
e
le
s
s
e
r
p
ix
e
l d
i
s
tr
ib
u
tio
n
ar
e
di
s
c
a
r
de
d f
r
om
t
h
e
s
ear
ch
.
O
t
h
er
es
s
en
t
i
al
i
n
f
o
r
m
at
i
o
n
r
eg
ar
d
i
n
g
t
h
e B
T
S
can
b
e
f
o
u
n
d
i
n
ear
l
i
er
w
o
r
k
[
2
0
]
,
[
21
]
.
I
n
t
hi
s
a
p
p
r
o
a
c
h
,
a
th
r
e
s
h
o
ld
li
m
it (
T
m
in
a
nd
T
ma
x
)
i
s
a
s
s
i
g
ne
d
d
ur
i
ng t
he
s
e
a
r
c
h,
w
h
ic
h
w
ill h
e
lp
to
a
tta
in
a
b
e
tte
r
r
e
s
ul
t
.
F
ur
t
he
r
,
a
m
u
l
tip
le
o
b
j
e
c
tiv
e
f
u
n
c
t
io
n
(
M
O
F
)
i
s
al
s
o
s
u
g
g
es
t
ed
w
i
t
h
p
ar
a
m
et
er
s
,
s
u
c
h
as
b
r
a
nc
h c
o
nt
r
o
l
va
l
ve
(
B
CV
)
,
p
eak
s
i
g
n
al
-
to
-
n
o
is
e
r
a
tio
(P
S
N
R
) a
n
d
s
t
r
uc
t
ur
a
l
-
s
im
il
a
r
i
ty
-
i
nd
e
x
-
m
eas
u
r
e (
S
SI
M
)
;
w
hi
c
h he
l
p
s
t
o
ach
i
ev
e a
n
en
h
an
ce r
es
u
l
t
c
o
m
p
ar
ed
t
o
t
h
e al
g
o
r
i
t
h
m
w
i
t
h
s
in
g
le
-
o
b
j
e
c
tiv
e
-
f
u
nc
t
i
o
n
(
SO
F)
.
T
h
i
s
w
or
k
a
l
s
o c
ons
i
de
r
s
t
h
e f
eat
u
r
e
-
s
i
m
ila
r
it
y
-
i
nd
e
x
(
F
S
I
)
[
3
3
]
to
te
s
t th
e
e
m
in
e
n
c
e
o
f
t
h
e
p
r
o
ces
s
ed
i
m
a
g
e.
T
h
e p
er
f
o
r
m
an
ce o
f
t
h
e p
r
o
p
o
s
ed
M
O
A
i
s
t
he
n va
l
i
d
a
t
e
d
a
ga
i
ns
t
t
he
o
t
he
r
he
ur
i
s
t
i
c
p
r
o
ced
u
r
es
,
s
u
ch
a
s
P
S
O
[1
6
],
b
a
c
te
r
ia
l f
o
r
a
g
i
n
g
o
p
ti
m
iz
a
tio
n
(B
F
O
) [3
4
]
,
[
3
5
],
F
A
[2
1
]
,
[
36]
,
b
a
t a
lg
o
r
ith
m
(B
A
) [2
7
],
cu
ck
o
o
s
ear
c
h
(
C
S
) [7
] a
n
d
mo
t
h
-
f
la
m
e
o
p
ti
m
i
z
a
tio
n
(M
F
O
)
[3
7
]
,
[
3
8
]
e
xi
s
t
i
ng
i
n t
he
l
i
t
e
r
a
t
ur
e
a
n
d
t
h
e
pe
r
f
or
m
a
n
c
e
of
t
h
e
M
O
A
on
t
h
e
I
M
L
T
pr
obl
e
m
i
s
v
a
l
i
da
t
e
d us
i
ng
t
h
e
p
-
va
l
ue
a
t
t
a
i
ne
d
us
i
n
g
t
he
W
i
l
co
x
o
n
r
an
k
t
e
s
t
.
T
h
e r
es
u
l
t
at
t
ai
n
ed
w
i
t
h
t
h
e p
r
o
p
o
s
ed
w
o
r
k
co
n
f
i
r
m
s
t
h
at
,
t
h
e M
O
A
ap
p
r
o
ach
h
el
p
s
t
o
ach
i
ev
e a b
et
t
er
r
es
u
l
t
o
n
t
h
e co
n
s
i
d
er
ed
t
es
t
i
m
a
g
e co
m
p
ar
e
d
t
o
t
h
e al
t
er
n
at
i
v
e al
g
o
r
i
t
h
m
s
.
2.
R
ES
EA
R
C
H
M
ETH
O
D
T
h
e s
u
p
er
i
o
r
i
t
y
o
f
al
l
t
h
e o
p
t
i
m
i
zat
i
o
n
p
r
o
ces
s
d
ep
en
d
s
o
n
t
h
e p
r
o
ced
u
r
e i
m
p
l
e
m
en
t
ed
an
d
t
h
i
s
s
ect
i
o
n
o
f
t
h
e r
es
ear
ch
d
ep
i
ct
s
t
h
e o
u
t
co
m
e at
t
ai
n
ed
w
i
t
h
M
O
A
.
2
.1
.
O
v
erv
i
e
w
o
f
p
ro
p
o
s
ed
m
et
h
o
d
T
h
e
i
m
p
le
m
e
n
ta
tio
n
o
f
M
O
A
a
s
s
is
te
d
I
M
L
T
is
d
e
p
ic
te
d
in
F
ig
u
r
e
1
.
T
h
e
ta
s
k
i
s
to
f
i
n
d
t
h
e
o
p
ti
m
a
l
t
h
r
e
s
h
ol
d ba
s
e
d on
t
h
e
a
s
s
i
gne
d t
h
r
e
s
h
ol
d b
y
m
a
xi
m
i
z
i
ng
t
h
e
M
O
F
.
I
n
t
h
i
s
ap
p
r
o
ach
,
t
h
e
M
O
A
i
s
i
n
i
t
i
al
i
s
ed
w
it
h
t
h
e
e
s
s
e
n
tia
l
p
a
r
a
m
e
te
r
s
a
n
d
a
llo
w
e
d
to
f
in
d
t
h
e
o
p
tim
a
l
th
r
e
s
h
o
ld
f
o
r
th
e
c
h
o
s
e
n
te
s
t
p
ic
tu
r
e
.
I
n
th
i
s
w
o
r
k
,
a
B
T
S
is
i
m
p
le
m
e
n
te
d
to
a
tta
in
a
f
in
e
s
t r
e
s
u
lt
w
ith
le
s
s
e
r
ite
r
a
tio
n
.
T
h
e
O
ts
u
is
c
o
n
s
id
e
r
e
d
a
s
th
e
m
e
t
h
odol
ogy
t
o e
nh
a
n
c
e
t
h
e
i
m
a
g
e
ba
s
e
d on t
h
r
e
s
h
ol
ds
.
T
h
e
e
s
s
e
n
t
i
a
l
i
n
f
or
m
a
t
i
on on
O
t
s
u c
a
n be
f
o
und
i
n
[
3
9
]
.
A
f
t
er
at
t
ai
n
i
n
g
t
h
e
r
es
u
l
t
,
a p
i
x
el
w
i
s
e co
m
p
ar
i
s
o
n
i
s
p
er
f
o
r
m
ed
b
et
w
een
t
h
e o
r
i
g
i
n
al
an
d
p
r
o
ces
s
ed
i
m
a
g
e an
d
t
h
e es
s
en
t
i
a
l
I
Q
P
ar
e co
m
p
u
t
ed
.
T
h
i
s
w
o
r
k
i
s
i
m
p
l
e
m
e
n
t
ed
w
i
t
h
a M
O
F
,
w
h
i
ch
al
s
o
h
el
p
s
t
o
ach
i
ev
e
a b
et
t
er
t
hr
e
s
ho
l
d
r
e
s
ul
t
f
o
r
e
ve
r
y
i
m
ag
e
co
n
s
i
d
er
ed
i
n
t
h
i
s
w
o
r
k
.
T
h
e en
t
i
r
e o
p
er
at
i
o
n
w
o
r
k
s
as
a
cl
o
s
ed
l
o
o
p
p
r
o
ces
s
an
d
co
n
t
i
n
u
ed
t
i
l
l
t
h
e f
i
n
es
t
t
h
r
es
h
o
l
d
i
s
r
each
ed
.
A
f
t
er
p
er
f
o
r
m
i
n
g
t
h
e es
s
e
n
t
i
al
o
p
er
at
i
o
n
,
t
h
e at
t
ai
n
ed
r
es
u
l
t
s
ar
e co
m
p
ar
ed
an
d
v
al
i
d
at
ed
b
as
ed
o
n
t
h
e co
m
p
u
t
ed
I
Q
P
.
2.
2.
B
e
n
c
h
m
ar
k
i
m
age
d
at
ab
as
e
A
co
n
s
i
d
er
ab
l
e n
u
m
b
er
o
f
R
G
B
/
g
r
a
y
s
ca
l
e i
m
a
g
es
ar
e av
ai
l
ab
l
e i
n
t
h
e
l
i
t
er
at
u
r
e [
4
0
]
-
[
42]
a
n
d t
hi
s
w
or
k
c
o
n
s
i
de
r
e
d t
h
e
c
o
m
m
o
n
l
y
us
e
d be
n
c
hm
a
r
k
i
m
a
g
e
s
of
d
i
m
e
n
s
i
o
n
512x
512x
1 f
or
t
h
e
e
x
pe
r
i
m
e
n
t
a
l
s
t
u
d
y
.
A
ll th
e
s
e
i
m
a
g
e
s
w
ill h
a
v
e
t
h
e
ir
o
w
n
p
i
x
e
l d
is
tr
ib
u
tio
n
,
r
a
n
g
in
g
f
r
o
m
s
m
o
o
t
h
to
c
o
m
p
le
x
a
n
d
th
e
s
a
m
p
le
te
s
t
i
m
a
g
es
co
n
s
i
d
er
ed
f
o
r
i
n
v
es
t
i
g
at
i
o
n
ar
e d
ep
i
ct
ed
i
n
F
i
g
u
r
e
2
.
E
v
er
y
i
m
a
g
e i
s
co
n
s
i
d
er
ed
f
o
r
t
h
e as
s
es
s
m
en
t
an
d
f
o
r
t
h
e d
em
o
n
s
t
r
at
i
o
n
,
t
h
e
r
es
u
l
t
at
t
ai
n
ed
f
o
r
B
ar
b
ar
a,
L
en
a (
s
m
o
o
th
p
ix
e
l d
is
tr
ib
u
tio
n
)
a
n
d
B
u
tte
r
f
l
y
,
B
e
e
(
c
o
m
p
l
e
x p
i
xe
l
d
i
s
t
r
i
b
ut
i
o
n)
s
ho
w
n i
n F
i
g
ur
e
3
i
s
p
r
e
s
e
nt
e
d
a
nd
d
i
s
c
u
s
s
e
d
i
n t
hi
s
r
e
s
e
a
r
c
h.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
,
V
o
l.
11
, N
o
.
6
,
D
ecem
b
er
2
0
2
1
:
5
420
-
54
29
5422
F
i
g
ur
e
1
.
V
a
r
i
ou
s
s
t
a
g
e
s
of
t
h
e
pr
op
os
e
d t
h
r
e
s
h
ol
di
ng
pr
oc
e
d
u
r
e
F
i
g
u
r
e 2
.
B
en
ch
m
ar
k
t
es
t
i
m
a
g
es
o
f
d
i
m
en
s
i
o
n
512x
512x
1
(
a)
(b
)
F
i
g
u
r
e 3
.
I
m
a
g
es
co
n
s
i
d
er
ed
f
o
r
i
n
v
es
t
i
g
at
i
o
n
,
(
a
)
s
m
o
o
th
p
i
x
e
l d
is
tr
ib
u
t
io
n
,
(
b
)
c
o
m
p
le
x
p
ix
e
l d
is
tr
ib
u
tio
n
2
.3
.
M
ay
f
l
y
op
t
i
m
i
z
at
i
on
al
gor
i
t
h
m
M
O
A
i
s
o
n
e o
f
t
h
e r
ecen
t
popu
l
a
t
i
on
-
b
as
ed
m
e
t
h
ods
i
nv
e
nt
e
d i
n
202
0 [
28
]
,
[
2
9
]
.
A
s
p
er
t
h
e ear
l
i
er
s
t
at
e
m
en
t
b
y
t
h
e au
t
h
o
r
s
,
t
h
e
M
O
A
i
s
d
er
i
v
ed
f
r
o
m
t
h
e P
S
O
an
d
i
t
co
m
b
i
n
es
al
l
t
h
e
m
aj
o
r
s
t
r
en
g
t
h
s
i
n
P
S
O
,
F
A
a
n
d G
A
.
T
h
e
c
o
n
c
e
pt
of
M
F
O
c
ons
i
s
t
t
h
e
f
ol
l
o
w
i
n
g ope
r
a
t
i
on
s
;
i
)
I
n
it
ia
liz
a
tio
n
o
f
e
q
u
a
l
n
u
m
b
e
r
o
f
ma
l
e
an
d
f
e
m
al
e
a
ge
nt
s
,
ii
)
A
l
l
o
w
i
ng t
he
m
a
le
-
M
a
yf
l
y t
o
i
d
e
nt
i
f
y t
he
G
be
st
f
o
r
t
he
c
ho
s
e
n t
a
s
k
,
iii
)
A
llo
w
i
n
g
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
I
S
S
N
:
2088
-
8708
I
m
age
m
ul
t
i
-
l
evel
-
th
r
e
s
h
o
ld
in
g
w
ith
M
a
y
fly
o
p
tim
iz
a
tio
n
(
Se
i
f
e
di
ne
K
adr
y
)
5423
fe
m
a
l
e
-
M
a
y
f
l
y
t
o
s
ear
ch
an
d
co
m
b
i
n
e
w
i
t
h
m
a
l
e
-
M
a
y
f
l
y
p
o
s
itio
n
e
d
a
t G
best
,
iv
)
O
f
f
s
p
r
i
n
g ge
ne
r
a
t
i
o
n a
nd
v
)
T
e
r
m
i
na
t
i
n
g t
he
s
e
a
r
c
h a
nd
d
i
s
p
l
a
y
i
ng t
he
f
i
na
l
r
e
s
ul
t
.
T
he
o
v
er
al
l
p
er
f
o
r
m
a
n
ce o
f
t
h
e M
O
A
d
ep
en
d
s
o
n
t
h
e i
n
i
t
i
al
p
o
s
i
t
i
o
n
o
f
t
h
e
m
a
l
e an
d
i
t
s
at
t
r
act
i
o
n
d
i
s
t
an
ce f
o
r
t
h
e f
e
m
al
e.
I
n
t
h
i
s
ap
p
r
o
ach
,
w
h
en
t
h
e ag
en
t
s
ar
e ar
b
i
t
r
a
r
i
l
y
i
n
i
t
i
al
i
zed
i
n
t
h
e
s
ear
ch
s
p
ace
w
i
t
h
eq
u
al
n
u
m
b
er
s
o
f
m
al
e
a
n
d
f
e
m
al
e
a
g
e
n
t
s
;
ev
er
y
M
a
y
f
l
i
es
ar
e
al
l
o
w
ed
t
o
co
n
v
er
g
e
t
o
w
ar
d
s
t
h
e
G
best
wi
t
h
i
n
cr
eas
e i
n
t
h
e co
n
v
er
g
e
n
ce.
T
h
i
s
p
r
o
ces
s
w
i
l
l
b
e s
t
o
p
p
ed
w
h
e
n
eq
u
al
n
u
m
b
er
s
o
f
o
f
f
s
p
r
i
n
g
s
i
s
g
en
er
at
ed
b
y
ev
er
y
p
ai
r
o
f
a
g
e
n
t
s
.
I
n
o
r
d
er
t
o
t
er
m
i
n
at
e
t
h
e p
r
o
ces
s
,
e
v
e
r
y
g
en
er
at
ed
o
f
f
s
p
r
i
n
g
s
ar
e
as
s
i
g
n
e
d
w
i
t
h
a zer
o
v
e
lo
c
it
y
,
s
o
th
a
t
it c
a
n
n
o
t
m
o
v
e
f
u
r
t
h
e
r
.
T
h
e
o
th
e
r
e
s
s
e
n
tia
l in
f
o
r
m
a
tio
n
c
a
n
b
e
f
o
u
n
d
in
[
2
8
]
,
[
29]
a
n
d i
t
s
b
as
i
c co
d
e can
b
e acces
s
ed
f
r
o
m
[
4
3
]
.
F
i
g
ur
e
4
gr
a
p
hi
c
a
l
l
y
d
e
p
i
c
t
s
t
he
w
o
r
ki
ng
m
e
c
ha
ni
s
m
o
f
t
he
M
O
A
w
i
t
h va
r
i
o
u
s
s
t
a
ge
s
,
s
uc
h a
s
Fi
g
ur
e
4
(
a
)
in
itia
liz
a
t
io
n
,
F
i
g
ur
e
4
(
b
)
is
id
e
n
tif
ic
a
tio
n
o
f
G
best
b
y
m
a
l
e
a
ge
nt
a
nd
F
i
g
u
re
4
(
c
)
d
e
p
ic
ts
th
e
m
o
ve
m
e
nt
o
f
f
e
m
a
l
e
t
o
w
a
r
d
t
he
m
a
l
e
a
nd
o
f
f
s
p
r
i
ng
ge
ne
r
a
t
i
o
n.
A
s
t
he
o
p
e
r
a
t
i
o
n
s
s
ho
w
n
i
n F
i
g
ur
e
4
,
t
he
m
at
h
e
m
a
t
i
cal
e
x
p
r
es
s
i
o
n
f
o
r
t
h
e M
O
A
is
al
s
o
h
a
v
e v
ar
i
o
u
s
p
h
as
es
a
nd
,
i
n t
hi
s
w
o
r
k,
th
e
a
lg
o
r
ith
m
p
a
r
a
m
e
te
r
s
c
on
s
i
de
r
e
d i
n
[
28
]
,
[
4
3
]
i
s
a
d
o
p
t
e
d
d
ur
i
ng t
he
i
m
p
l
e
m
e
nt
a
t
i
o
n.
F
ur
t
he
r
,
t
hi
s
a
l
go
r
i
t
h
m
c
o
nt
a
i
ns
t
he
b
a
s
i
c
o
p
e
r
a
t
i
o
n o
f
P
S
O
a
nd
F
A
d
ur
i
ng t
he
m
o
ve
m
e
nt
a
nd
e
ve
r
y
a
ge
nt
(
M
a
yf
l
y
)
l
o
c
a
t
i
o
n c
a
n
b
e
v
ar
i
ed
b
as
ed
o
n
t
h
e
p
o
s
itio
n
a
n
d
v
e
lo
c
it
y
e
q
u
a
tio
n
p
r
e
s
e
n
t in
t
h
e
e
a
r
lie
r
a
lg
o
r
ith
m
s
.
(
a)
(b
)
(
c)
F
i
g
ur
e
4
.
V
a
r
i
o
us
s
t
a
ge
s
o
f
t
he
M
a
y
f
l
y
a
l
go
r
i
t
h
m
,
(a
)
in
itia
l
M
a
y
f
l
y
p
o
s
itio
n
,
(
b
)
f
ir
s
t s
e
a
r
c
h
b
y
m
a
le
-
M
a
yf
l
y,
(c
)
f
i
n
al
s
ear
c
h
b
y
fe
m
a
l
e
-
M
a
yf
l
y
L
et
t
h
er
e ex
i
s
t
s
eq
u
al
n
u
m
b
er
o
f
m
al
e (
M)
an
d
f
e
m
al
e (
F
)
M
a
y
f
l
y
i
n
a d
-
d
i
m
en
s
i
o
n
al
s
ear
c
h
l
o
cat
i
o
n
a
nd
t
he
t
o
t
a
l
nu
m
b
e
r
o
f
a
ge
nt
s
(
M
a
yf
l
ys
)
a
r
e
d
e
p
i
c
t
e
d
by i
=
1
,
2
…
,
N
.
(
N
=
2
0
i
s
a
s
s
i
gne
d
i
n t
hi
s
w
o
r
k)
.
D
u
r
i
n
g
t
h
e o
p
t
i
m
i
zat
i
o
n
s
ear
c
h
,
each
ag
e
n
t is
r
a
n
d
o
m
l
y
in
i
tia
liz
e
d
in
th
e
s
e
a
r
c
h
lo
c
a
lit
y
a
n
d
e
v
e
r
y
a
g
e
n
t is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
,
V
o
l.
11
, N
o
.
6
,
D
ecem
b
er
2
0
2
1
:
5
420
-
54
29
5424
a
llo
w
e
d
to
m
o
v
e
to
w
a
r
d
s
th
e
f
i
n
e
s
t p
o
s
itio
n
(
G
best
)
,
w
h
en
t
h
e i
t
er
at
i
o
n
i
n
cr
eas
e
s
.
A
s
i
n
F
i
g
ur
e
4
(
b
)
,
t
h
e
m
al
e
i
s
al
l
o
w
ed
t
o
r
each
t
h
e
G
best
by
a
dj
u
s
t
i
ng
i
t
s
pos
i
t
i
on
a
n
d
v
e
l
oc
i
t
y
.
T
h
e
m
o
v
e
m
en
t
o
f
t
h
e ag
en
t
t
o
w
ar
d
s
t
h
e
f
i
n
a
l d
e
s
ti
n
a
tio
n
w
i
ll b
e
g
u
id
e
d
b
y
t
h
e
C
a
r
te
s
ia
n
d
is
ta
n
c
e
a
n
d
th
e
in
c
r
e
a
s
e
d
ite
r
a
tio
n
.
T
h
is
o
p
e
r
a
tio
n
is
s
i
m
ila
r
t
o t
h
e
F
A
di
s
c
u
s
s
e
d i
n
[
21]
.
T
h
e
pos
i
t
i
on
a
n
d t
h
e
v
e
l
oc
i
t
y
upda
t
e
e
qu
a
t
i
on
s
a
r
e
s
h
o
w
n i
n
(
1)
a
n
d (
2)
;
1
t
i
V
t
i
M
1
t
i
M
+
+
=
+
(1
)
)
t
j
,
i
M
bes
t
G
(
2
D
e
*
2
C
)
t
j
,
i
M
bes
t
P
(
2
D
e
*
1
C
t
j
,
i
V
1
t
j
,
i
V
j
,
i
g
j
,
i
p
−
β
−
+
−
β
−
+
=
+
(2
)
w
h
er
e
a
nd
+
1
a
r
e
in
itia
l a
n
d
m
o
d
if
ie
d
p
o
s
it
io
n
s
,
+
1
a
nd
,
+
1
in
itia
l
a
n
d
m
o
d
if
ie
d
v
e
lo
c
itie
s
r
es
p
ect
i
v
el
y
,
p
er
s
o
n
al
l
ear
n
i
n
g
p
ar
a
m
et
er
(
1
C
)
=1
,
P
er
s
o
n
al
l
ear
n
i
n
g
p
ar
a
m
e
t
er
(
2
C
)
=
1
.5
,
β
=2
an
d
p
D
a
nd
g
D
ar
e C
ar
t
es
i
an
d
i
s
t
an
ce.
In
(
2
)
i
s
f
r
a
m
e
d
b
y
c
o
m
b
i
ni
n
g t
he
F
A
a
nd
P
S
O
va
l
ue
s
.
W
he
n t
he
up
d
a
t
i
o
n c
o
nt
i
n
ue
s
b
a
s
e
d
t
he
p
r
o
gr
e
s
s
i
o
n i
n
i
t
e
r
at
i
o
n
s
,
ev
er
y
m
al
e (
M)
w
i
l
l
r
each
t
h
e
bes
t
G
an
d
p
er
f
o
r
m
s
a
v
el
o
ci
t
y
u
p
d
at
i
o
n
t
o
at
t
r
act
t
h
e
f
e
m
al
e (
F
)
b
y
p
er
f
o
r
m
i
n
g
a u
n
i
q
u
e n
u
p
t
i
al
d
an
ce
(
m
o
v
i
n
g
u
p
an
d
d
o
w
n
o
n
a
w
a
t
er
s
u
r
f
ace)
.
T
h
e v
el
o
ci
t
y
u
p
d
at
e d
u
r
i
n
g
t
h
i
s
p
r
o
ces
s
can
b
e d
ef
i
n
e
d
a
s
;
R
*
d
t
j
,
i
V
1
t
j
,
i
V
+
=
+
(3
)
w
h
er
e n
u
p
t
i
al
d
an
ce
v
al
u
e (
d
)
=5
an
d
R
= r
an
d
o
m
n
u
m
er
al
[
-
1
],
[
1]
.
A
f
t
er
t
h
e o
p
t
i
m
a
l
s
ear
c
h
b
y
m
al
e i
s
co
m
p
l
e
t
ed
,
ev
er
y
f
e
m
al
e (
F
)
i
s
t
h
en
al
l
o
w
ed
t
o
i
d
en
t
i
f
y
a
m
al
e,
w
ho i
s
a
t
bes
t
G
.
Fi
g
ur
e
4
(
c)
d
e
p
i
ct
s
t
h
i
s
p
r
o
ces
s
i
n
w
h
i
c
h
a f
e
m
a
l
e
(
F
)
m
a
y
m
o
v
e t
o
w
ar
d
s
t
h
e
m
al
e b
as
ed
o
n
t
h
e d
i
s
t
an
ce (
)
o
r
i
t
m
a
y
e
s
ca
p
e t
o
a n
e
w
l
o
cat
i
o
n
u
s
i
n
g
a r
an
d
o
m
-
wa
l
k
(
W
=1
)
v
al
u
e.
T
h
e
m
at
h
e
m
at
i
cal
e
x
pr
e
s
s
i
on
f
or
pos
i
t
i
on
a
n
d
v
e
l
oc
i
t
y
u
pda
t
e
f
or
F
i
s
de
pi
c
t
e
d be
l
o
w,
1
t
i
V
t
i
F
1
t
i
F
+
+
=
+
(4
)
)
i
O(
M
)
i
O(
F
if
r
*
W
t
j
,
i
V
)
i
O(
M
)
i
O(
F
if
)
t
j
,
i
Y
t
j
,
i
X
(
D
e
2
C
t
j
,
i
V
1
t
j
,
i
V
2
mf
≤
+
>
−
β
−
+
=
+
(5
)
w
h
er
e
O
=
m
a
x
i
m
i
zed
o
b
j
ect
i
v
e v
al
u
e.
W
h
en
t
h
e i
t
er
at
i
o
n
i
n
cr
eas
es
,
ev
er
y
F
w
i
l
l
r
each
t
h
e ap
p
r
o
p
r
i
at
e M
an
d
t
h
e o
f
f
s
p
r
i
n
g
g
e
n
er
at
i
o
n
t
a
ke
p
l
ace
a
s
d
e
p
ic
te
d
in
F
ig
ur
e
4(
c
)
.
I
n
t
h
e
M
O
A
,
t
h
e
num
be
r
of
of
f
s
pr
i
ng
s
i
s
e
qu
a
l
t
o t
h
e
c
om
bi
n
e
d M
a
n
d F
.
H
en
ce,
ev
er
y
m
at
i
n
g
w
i
l
l
r
es
u
l
t
i
n
a M an
d
F
o
f
f
s
p
r
i
n
g
w
i
t
h
an
i
n
i
t
i
al
v
el
o
ci
t
y
v
al
u
e o
f
Z
er
o
.
F
o
r t
h
e
I
M
L
T
p
r
o
b
l
em
,
o
n
l
y
t
h
e s
ear
c
h
p
er
f
o
r
m
a
n
ce o
f
M
an
d
F
al
o
n
e co
n
s
i
d
er
ed
an
d
t
he
s
e
a
r
c
h b
y
t
he
o
f
f
s
p
r
i
ng i
s
i
g
no
r
e
d
b
y
n
u
lli
f
y
i
n
g
its
in
itia
l
v
e
lo
c
i
t
y
o
p
e
r
a
to
r
.
F
r
o
m
t
h
e
a
b
o
v
e
d
is
c
u
s
s
e
d
e
q
u
a
tio
n
s
,
i
t is
c
le
a
r
th
a
t,
t
h
e
p
r
o
p
o
s
e
d
M
O
A
i
s
f
o
r
m
ed
b
y
co
m
b
i
n
i
n
g
t
h
e f
i
n
e
s
t
f
eat
u
r
es
o
f
t
h
e F
A
,
P
S
O
an
d
F
A
.
O
t
h
er
r
el
at
ed
i
n
f
o
r
m
at
i
o
n
can
b
e
obt
ai
n
ed
f
r
o
m
[
2
8
]
,
[
29
]
.
2
.4
.
O
b
je
c
ti
v
e
fu
n
c
ti
o
n
T
h
e o
v
er
al
l
p
er
f
o
r
m
an
ce
of
t
h
e
pr
opos
e
d
opt
i
m
i
z
a
t
i
o
n
t
a
s
k
de
pe
n
ds
on
t
h
e
c
h
os
e
n
obj
e
c
t
i
v
e
v
a
l
u
e
a
n
d i
n
t
h
i
s
w
or
k
m
a
x
i
m
i
z
a
t
i
on pr
obl
e
m
i
s
c
ons
i
de
r
e
d ba
s
e
d
on
(
6)
;
SSI
M
*
3
w
PSNR
*
2
w
BCV
*
1
w
O
M
aximize
+
+
=
=
(6
)
w
h
er
e
1
w
1
=
,
5
.
0
w
w
3
2
=
=
T
he
(6
)
p
r
e
s
e
n
ts
a
w
e
ig
h
te
d
s
u
m
o
f
o
b
j
e
c
tiv
e
f
u
n
c
tio
n
w
it
h
v
a
r
io
u
s
w
e
i
g
h
ts
.
I
n
th
i
s
w
o
r
k
,
B
C
V
is
gi
ve
n
t
he
hi
ghe
r
p
r
i
o
r
i
t
y
a
nd
P
S
N
R
a
nd
S
S
I
M
a
r
e
gi
ve
n
w
i
t
h
e
q
ua
l
p
r
i
o
r
i
t
i
e
s
.
T
he
t
a
s
k
o
f
t
he
M
O
A
i
s
t
o
e
x
pl
or
e
t
h
e
boun
de
d t
h
r
e
s
h
ol
d v
a
l
u
e
a
n
d i
de
nt
i
f
y
t
h
e
f
i
n
e
s
t
t
h
r
e
s
h
ol
d c
o
m
bi
n
a
t
i
on
,
w
hi
c
h
pr
ov
i
de
s
t
he
O
ma
x
.
O
th
e
r
r
e
la
te
d
in
f
o
r
m
a
tio
n
r
e
g
a
r
d
in
g
t
h
e
w
e
ig
h
te
d
s
u
m
o
f
M
O
F
c
a
n
be
f
oun
d i
n
[
20]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
I
S
S
N
:
2088
-
8708
I
m
age
m
ul
t
i
-
l
evel
-
th
r
e
s
h
o
ld
in
g
w
ith
M
a
y
fly
o
p
tim
iz
a
tio
n
(
Se
i
f
e
di
ne
K
adr
y
)
5425
2
.5
.
I
m
age
-
qua
l
i
t
y
-
p
a
ra
m
et
ers
A
ss
e
s
s
m
e
n
t
o
f
at
t
ai
n
ed
I
Q
P
i
s
e
s
s
e
n
t
i
al
t
o
co
n
f
i
r
m
t
h
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
i
m
a
g
e
p
r
o
ces
s
i
n
g
a
p
p
r
o
a
c
h.
I
n
t
hi
s
w
o
r
k,
t
he
we
l
l
-
k
n
o
wn
I
Q
P
s,
s
u
c
h
a
s
R
M
S
E
,
P
S
N
R,
S
S
I
M
,
N
A
E
,
N
CC,
A
D
a
n
d
S
C
a
r
e
co
n
s
i
d
er
ed
.
T
h
e o
t
h
er
i
m
a
g
e
q
u
al
i
t
y
m
ea
s
u
r
e cal
l
ed
t
h
e
f
e
a
tu
r
e
s
i
m
ila
r
i
t
y
in
d
e
x
(
F
S
I
M
)
[
33]
i
s
a
l
s
o a
dopt
e
d
i
n
t
h
i
s
w
o
r
k
t
o
co
n
f
i
r
m
t
h
e em
i
n
en
ce o
f
t
h
e p
r
o
p
o
s
ed
w
o
r
k
.
T
h
e m
at
h
e
m
at
i
cal
ex
p
r
es
s
i
o
n
f
o
r
t
h
e co
n
s
i
d
er
ed
I
Q
P
s
c
a
n b
e
f
o
und
i
n [
4
4
]
,
[
4
5]
.
2
.6
.
O
t
h
er a
l
g
o
ri
t
h
m
s
T
o
v
al
i
d
at
e t
h
e p
er
f
o
r
m
a
n
ce o
f
t
h
M
O
A
,
t
h
e
we
l
l
-
k
n
o
wn
he
ur
i
s
t
i
c
m
e
t
ho
d
s
,
s
uc
h a
s
P
S
O
,
B
F
O
,
B
A
,
F
A
,
C
S
an
d
M
F
O
i
s
co
n
s
i
d
er
ed
an
d
t
h
e p
er
f
o
r
m
an
ce o
f
t
h
es
e
m
et
h
o
d
s
ar
e v
al
i
d
at
ed
b
y
co
m
p
u
t
i
n
g
t
h
e
at
t
ai
n
ed
O
t
s
u
’
s
B
C
V
,
co
n
v
er
g
en
ce
r
at
e a
n
d
t
h
e W
i
l
co
x
o
n
r
an
k
t
es
t
.
F
o
l
l
o
w
i
n
g
al
g
o
r
i
t
h
m
p
ar
a
m
et
er
s
ar
e
c
o
mmo
n
l
y
a
s
s
i
g
n
e
d f
or
e
a
c
h
a
l
g
or
i
t
hm
;
n
um
be
r
of
a
g
e
n
t
s
=
20,
s
e
a
r
c
h
di
m
e
ns
i
on
=
T
(
i
e
.
T
=
2 t
o 5)
,
m
a
x
i
m
um
ite
r
a
tio
n
s
3
0
0
0
a
n
d
th
e
s
to
p
p
i
n
g
c
r
ite
r
ia
=
m
a
x
i
m
a
l ite
r
a
tio
n
o
r
o
p
ti
m
iz
e
d
r
e
s
u
lt.
F
o
r
e
a
c
h
im
a
g
e
,
f
i
v
e
tr
ia
ls
a
r
e
i
m
p
l
e
m
e
nt
e
d
w
i
t
h
e
a
c
h
T
a
nd
t
he
i
m
a
ge
w
h
i
c
h
gi
ve
s
t
he
f
i
n
es
t
I
Q
P
i
s
co
n
s
i
d
er
ed
as
t
h
e
t
hr
e
s
ho
l
d
i
ma
g
e
.
O
t
h
e
r
e
s
s
e
nt
i
a
l
i
nf
or
m
a
t
i
o
n
r
e
g
a
r
di
ng
t
h
e
i
m
p
l
e
m
e
n
t
a
t
i
on
o
f
t
h
e
a
l
g
or
i
t
hm
s
f
or
I
M
L
T
pr
obl
e
m
c
a
n
be
f
oun
d
i
n [
4
4
]
-
[
47]
.
3.
R
ES
U
LT
S
AND
D
I
SC
U
SS
I
O
N
T
h
is
s
e
c
tio
n
p
r
e
s
e
n
t
s
th
e
e
x
p
e
r
i
m
e
n
ta
l r
e
s
u
lts
a
n
d
it
s
d
is
c
u
s
s
io
n
s
.
A
ll t
h
e
e
x
p
e
r
i
m
e
n
ta
l i
n
v
e
s
ti
g
a
tio
n
i
s
i
m
p
l
e
m
e
n
t
ed
u
s
i
n
g
M
A
T
L
A
B
an
d
t
h
e p
er
f
o
r
m
an
ce o
f
t
h
e p
r
o
p
o
s
ed
t
ech
n
i
q
u
e i
s
co
n
f
i
r
m
ed
u
s
i
n
g
t
h
e
p
-
v
al
u
e at
t
ai
n
ed
w
i
t
h
t
h
e
W
ilc
o
x
o
n
te
s
t.
E
ve
r
y
i
m
a
ge
c
o
n
s
id
e
r
e
d
f
o
r
th
e
in
v
e
s
ti
g
a
tio
n
is
i
n
i
tia
ll
y
e
x
p
lo
r
e
d
a
n
d
a
t
hr
e
s
ho
l
d
’
s
s
e
a
r
c
h b
o
u
nd
a
r
y
i
s
a
s
s
i
g
ne
d
w
i
t
h a
m
i
ni
m
u
m
(
T
m
i
n)
a
nd
m
a
xi
m
u
m
(
T
m
a
x)
va
l
ue
a
nd
t
he
M
O
A
is
th
e
n
a
llo
w
e
d
to
e
x
p
lo
r
e
th
e
th
r
e
s
h
o
ld
b
o
u
n
d
a
r
y
to
f
i
n
d
th
e
f
in
e
s
t t
h
r
e
s
h
o
ld
.
T
h
e
s
a
m
p
le
th
r
e
s
h
ol
d bou
n
da
r
y
a
s
s
i
g
n
e
d
f
o
r
th
e
B
a
r
b
a
r
a
p
ic
tu
r
e
is
d
e
p
ic
te
d
in
F
i
g
u
r
e
5
.
I
n
t
h
is
w
o
r
k
,
th
e
o
p
ti
m
iz
a
tio
n
p
r
o
c
e
s
s
w
i
ll h
e
lp
to
f
i
nd
t
he
t
hr
e
s
ho
l
d
b
y
c
o
ns
i
d
e
r
i
ng;
T
m
i
n<
T
hr
e
s
ho
l
d
va
l
ue
s
<
T
m
a
x.
T
he
M
O
A
i
s
i
ni
t
i
a
l
i
z
e
d
w
i
t
h
t
he
p
ar
am
et
er
s
as
d
i
s
cu
s
s
ed
i
n
s
ec
t
i
o
n
2
.
3 a
n
d t
h
e
r
e
s
ul
t
s
a
t
t
a
i
n
e
d f
or
a
s
s
i
gn
e
d t
h
r
e
s
h
ol
ds
i
s
de
pi
c
t
e
d i
n
F
i
gu
r
e
6
a
lo
n
g
w
it
h
t
h
e
o
p
ti
m
a
l t
h
r
e
s
h
o
ld
a
tta
in
e
d
.
F
i
g
ur
e
5
.
F
i
xi
ng t
he
t
hr
e
s
ho
l
d
b
o
und
b
a
s
e
d
o
n t
he
p
i
xe
l
l
e
ve
l
Fr
o
m
Fi
g
ur
e
6
(
a
)
to
6
(
d
)
d
e
p
ic
ts
th
e
r
e
s
u
lts
a
tta
i
n
e
d
w
it
h
a
c
h
o
s
e
n
t
hr
e
s
ho
l
d
s
a
nd
t
hi
s
i
m
a
ge
c
o
nf
i
r
m
s
t
ha
t
;
b
a
s
e
d
o
n
t
he
t
hr
e
s
ho
l
d
va
l
ue
,
t
he
q
ua
l
i
t
y
a
nd
t
he
i
n
f
o
r
m
a
t
i
o
n c
o
nt
e
n
t
o
f
t
he
i
m
a
ge
va
r
i
e
s
co
n
s
i
d
er
ab
l
y
.
T
h
i
s
p
i
ct
u
r
e al
s
o
d
ep
i
ct
s
t
h
e
v
al
u
e o
f
t
h
e S
S
I
M
-
M
a
p,
w
hi
c
h c
onf
i
r
m
s
t
h
a
t
t
h
e
pr
opos
e
d
m
e
t
h
od
h
e
lp
s
to
a
t
ta
in
a
be
t
t
e
r
r
e
s
u
l
t
w
i
t
h
t
h
e
pr
opos
e
d m
e
t
h
odol
ogy
.
S
i
m
i
l
a
r
m
e
t
h
odol
ogy
i
s
i
m
p
l
e
m
e
n
t
e
d
f
or
ot
h
e
r
i
m
a
ge
s
a
nd
t
he
r
e
s
ul
t
s
a
t
t
a
i
ne
d
a
r
e
c
l
e
a
r
l
y
s
ho
w
n
i
n
F
i
gur
e
7
f
o
r
t
hr
e
s
ho
l
d
s
T
=
2 t
o 5
.
A
f
te
r
c
o
lle
c
ti
n
g
t
h
e
n
eces
s
ar
y
m
u
l
t
i
-
t
h
r
es
h
o
l
d
i
m
a
g
es
,
a co
m
p
ar
at
i
v
e as
s
es
s
m
e
n
t
is
th
e
n
i
m
p
le
m
e
n
te
d
b
e
t
w
e
e
n
th
e
o
r
ig
in
a
l i
m
a
g
e
an
d
t
h
e p
r
o
ces
s
ed
i
m
a
g
e a
n
d
t
h
e e
s
s
e
n
t
i
al
I
Q
P
d
ep
i
ct
ed
i
n
T
ab
l
e 1
i
s
co
m
p
u
t
ed
.
T
h
e
r
es
u
l
t
o
f
t
h
i
s
t
ab
l
e
co
n
f
i
r
m
s
t
h
at
,
w
h
e
n
t
h
e t
h
r
es
h
o
l
d
l
ev
el
i
n
cr
eas
e
s
,
t
h
e
p
r
i
m
e
p
ar
am
et
er
s
,
s
u
c
h
as
P
S
N
R
,
S
S
I
M
,
F
S
I
M an
d
NC
C
i
m
p
r
o
v
es
a
n
d
o
t
h
er
p
ar
am
et
er
s
,
s
u
c
h
as
r
oot
m
e
a
n
s
q
ua
r
e
d e
r
r
or
(
RM
S
E
)
,
N
AE
,
AD
a
n
d
s
i
n
g
l
e cr
o
ch
et
(
SC
)
d
ecr
eas
es
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
,
V
o
l.
11
, N
o
.
6
,
D
ecem
b
er
2
0
2
1
:
5
420
-
54
29
5426
F
i
g
ur
e
6
.
T
hr
e
s
ho
d
i
ng r
e
s
ul
t
a
t
t
a
i
ne
d
f
o
r
B
a
r
b
a
r
a
i
m
a
ge
F
i
g
u
r
e 7
.
R
es
u
l
t
ach
i
e
v
ed
f
o
r
t
h
e t
es
t
i
m
a
g
es
f
o
r
,
(a
)
T
=
2
,
(b
)
T
=3
,
(
c
)
T
=4
,
(
d
)
T
=5
T
ab
l
e 1
.
O
b
ta
in
e
d
I
Q
P
w
ith
t
h
e
M
O
A
a
s
si
st
e
d
I
M
L
T
Im
a
g
e
T
R
M
SE
PSN
R
(
d
B
)
S
S
I
M
F
S
IM
N
AE
NC
C
AD
SC
B
a
rb
a
ra
2
6
6
.
54
0
4
1
1
.
66
9
1
0
.
4
03
9
0
.
6
81
0
0
.
4
98
2
0
.
5
75
3
5
8
.
48
0
5
2
.
4
11
4
3
4
2
.
47
6
7
1
5
.
56
7
8
0
.
6
18
8
0
.
7
94
4
0
.
3
14
7
0
.
7
39
7
3
6
.
94
7
0
1
.
7
03
7
4
3
5
.
17
4
7
1
7
.
20
6
2
0
.
6
83
3
0
.
8
42
4
0
.
2
62
7
0
.
7
98
2
3
0
.
83
3
7
1
.
4
91
9
5
2
9
.
04
6
1
1
8
.
86
9
0
0
.
7
39
6
0
.
8
75
0
0
.
2
10
8
0
.
8
41
1
2
4
.
74
8
6
1
.
3
65
2
Le
n
a
2
5
9
.
33
0
1
1
2
.
66
5
3
0
.
3
74
1
0
.
6
52
2
0
.
5
14
2
0
.
5
55
1
5
0
.
92
7
4
2
.
5
68
4
3
4
2
.
58
0
5
1
5
.
54
6
6
0
.
5
47
4
0
.
7
37
8
0
.
3
76
0
0
.
6
92
4
3
7
.
24
2
2
1
.
8
91
9
4
3
0
.
26
7
4
1
8
.
51
1
3
0
.
6
83
2
0
.
8
12
3
0
.
2
66
4
0
.
7
83
3
2
6
.
39
0
5
1
.
5
64
4
5
2
3
.
87
7
7
2
0
.
57
0
9
0
.
7
54
9
0
.
8
63
9
0
.
2
08
7
0
.
8
32
6
2
0
.
67
3
1
1
.
4
07
5
B
u
tte
r
f
ly
2
7
2
.
59
7
0
1
0
.
91
2
4
0
.
3
52
4
0
.
6
48
2
0
.
5
50
6
0
.
5
36
7
6
4
.
28
8
9
2
.
4
10
4
3
5
1
.
50
7
7
1
3
.
89
3
4
0
.
5
42
1
0
.
7
28
0
0
.
3
71
0
0
.
7
05
2
4
3
.
32
6
2
1
.
7
17
5
4
3
6
.
67
8
8
1
6
.
84
2
5
0
.
6
87
7
0
.
7
90
5
0
.
2
51
2
0
.
7
99
4
2
9
.
33
0
9
1
.
4
57
6
5
3
7
.
36
1
9
1
6
.
68
2
2
0
.
6
92
0
0
.
8
15
7
0
.
2
42
3
0
.
8
26
1
2
8
.
29
0
1
1
.
3
46
6
B
ee
2
6
6
.
81
5
9
1
1
.
63
3
2
0
.
2
95
8
0
.
7
05
3
0
.
5
78
7
0
.
5
26
9
6
1
.
00
8
4
2
.
6
20
2
3
4
6
.
25
4
3
1
4
.
82
7
8
0
.
4
95
3
0
.
8
10
8
0
.
3
94
9
0
.
7
03
2
4
1
.
63
0
2
1
.
7
74
7
4
4
0
.
25
6
2
1
6
.
03
4
2
0
.
5
55
1
0
.
8
49
1
0
.
3
33
8
0
.
7
68
0
3
5
.
18
4
9
1
.
5
26
8
5
3
4
.
34
2
7
1
7
.
41
4
1
0
.
6
37
2
0
.
8
49
3
0
.
2
74
6
0
.
7
98
0
2
8
.
94
7
2
1
.
4
65
1
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
I
S
S
N
:
2088
-
8708
I
m
age
m
ul
t
i
-
l
evel
-
th
r
e
s
h
o
ld
in
g
w
ith
M
a
y
fly
o
p
tim
iz
a
tio
n
(
Se
i
f
e
di
ne
K
adr
y
)
5427
F
i
g
ur
e
8
d
ep
i
ct
s
t
h
e
m
a
x
i
m
al
v
al
u
e
o
f
O
r
s
u
’
s
B
C
V
o
b
t
ai
n
ed
f
o
r
B
ar
b
ar
a
i
m
ag
e
f
o
r
T
=5
an
d
t
h
i
s
f
i
g
u
r
e a
l
s
o
d
ep
i
ct
s
t
h
e
co
n
v
er
g
en
ce
o
f
t
h
e
al
g
o
r
i
t
h
m
t
o
w
ar
d
s
t
h
e
o
p
t
i
m
al
v
al
u
e.
T
h
e
M
O
A
h
e
lp
s
to
a
tta
i
n
a
m
ax
i
m
i
zed
B
C
V
co
m
p
ar
ed
t
o
o
t
h
er
m
et
h
o
d
s
.
F
u
r
t
h
er
,
t
h
e M
O
A
s
h
o
w
s
a
b
et
t
er
an
d
s
t
ead
y
co
n
v
er
g
e
n
ce
t
o
w
ar
d
s
t
h
e
m
ax
i
m
i
zed
B
C
V
.
T
h
i
s
co
n
f
i
r
m
s
t
h
at
t
h
e s
ear
c
h
p
er
f
o
r
m
a
n
ce o
f
M
O
A
i
s
g
o
o
d
co
m
p
ar
ed
t
o
o
t
h
er
m
et
h
o
d
s
co
n
s
i
d
er
ed
i
n
t
h
i
s
r
es
ear
ch
.
T
h
e
s
ta
tis
tic
a
l s
i
g
n
i
f
ic
a
n
c
e
o
f
th
e
M
O
A
w
it
h
r
e
s
p
e
c
t to
o
th
e
r
m
e
th
o
d
s
is
v
a
lid
a
te
d
u
s
i
n
g t
he
W
i
l
co
x
o
n
r
an
k
t
e
s
t
an
d
t
h
e at
t
ai
n
ed
r
es
u
l
t
s
ar
e d
ep
i
ct
ed
i
n
T
a
b
l
e 2
.
T
h
i
s
t
es
t
i
s
t
o
ev
al
u
at
e
t
h
e I
M
L
T
p
er
f
o
r
m
an
ce
o
f
t
h
e
M
O
A
w
i
t
h
o
t
h
er
co
n
s
i
d
er
ed
m
et
h
o
d
s
an
d
t
h
e
p
r
es
en
t
ed
r
es
u
l
t
co
n
f
i
r
m
s
t
h
at
t
h
e
p
-
v
al
u
e at
t
ai
n
ed
b
y
t
h
i
s
te
s
t is
<
0.
05,
w
h
i
c
h c
onf
i
r
m
s
t
h
e p
er
f
o
r
m
a
n
ce o
f
t
h
e M
O
A
i
s
b
et
t
e
r
o
n
t
h
e co
n
s
i
d
er
ed
t
h
r
e
s
h
ol
di
ng
pr
obl
e
m
[
48]
.
F
i
g
ur
e
8
.
C
o
m
p
ar
i
s
o
n
o
f
t
h
e s
ear
ch
t
r
aces
m
ad
e b
y
h
eu
r
i
s
t
i
c
al
g
o
r
i
t
h
m
s
f
o
r
B
ar
b
ar
a i
m
ag
e
w
i
t
h
T
=5
T
ab
l
e
2
.
P
e
r
f
o
r
m
a
nc
e
va
l
i
d
a
t
i
o
n o
f
M
O
A
w
i
t
h o
t
he
r
m
e
t
ho
d
s
us
i
n
g
W
i
l
c
o
xo
n r
a
nk t
e
s
t
A
l
g
o
r
ith
m
p
-
v
a
lu
e
M
O
A
v
e
r
s
u
s
PSO
0
.
0
01
5
M
O
A
v
e
r
su
s B
F
O
1
.
0
1
4
E
-
7
M
O
A
v
e
r
su
s F
A
0
.
0
02
7
M
O
A
v
e
r
su
s B
A
2
.
0
8
1
6
E
-
6
M
O
A
v
e
r
su
s C
S
0
.
0
8
3
6
E
-
7
M
O
A
v
e
r
su
s M
F
O
3
.
0
1
9
7
E
-
8
4.
CO
NCL
U
S
I
O
N
T
h
i
s
r
es
ear
ch
i
m
p
l
e
m
e
n
t
s
a
M
O
A
b
as
ed
I
M
L
T
f
o
r
a cl
as
s
o
f
b
en
c
h
m
ar
k
g
r
a
y
s
cal
e
i
m
a
g
es
u
s
i
n
g
a
n
o
v
el
o
b
j
ect
i
v
e f
u
n
ct
i
o
n
.
T
h
e
M
O
A
h
el
p
s
t
o
ach
i
e
v
e a
n
o
p
t
i
m
al
r
e
s
u
l
t
d
u
e
t
o
a
t
w
o
-
l
e
v
el
s
ear
ch
p
r
o
ces
s
an
d
p
r
o
v
id
e
s
b
e
tte
r
c
o
n
tr
a
s
t to
P
S
O
a
n
d
F
A
e
x
is
t
in
g
i
n
t
h
e
li
t
e
r
a
t
ur
e
.
I
n t
hi
s
w
o
r
k,
t
he
I
M
L
T
i
s
i
m
p
l
e
m
e
nt
e
d
f
o
r
T
=2
t
o
5
an
d
f
o
r
ex
ch
cas
e,
f
i
v
e t
r
i
al
s
ar
e i
m
p
l
e
m
e
n
t
ed
an
d
t
h
e b
es
t
r
es
u
l
t
a
m
o
n
g
t
h
e t
r
i
al
s
co
n
s
i
d
er
ed
as
t
h
e
o
p
tim
iz
e
d
r
e
s
u
lt.
A
f
te
r
g
e
tti
n
g
t
h
e
p
r
o
c
e
s
s
e
d
i
m
a
g
e
,
a
p
ix
e
l le
v
e
l c
o
m
p
a
r
is
o
n
w
i
th
t
h
e
o
r
ig
in
a
l te
s
t i
m
a
g
e
i
s
t
h
en
p
er
f
o
r
m
ed
a
n
d
t
h
e e
s
s
e
n
t
i
al
I
Q
P
s
ar
e co
m
p
u
t
ed
.
T
h
e o
b
t
ai
n
ed
r
es
u
l
t
o
f
t
h
i
s
s
t
u
d
y
co
n
f
i
r
m
s
t
h
a
t
t
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ach
h
el
p
s
t
o
g
et
b
et
t
er
o
u
t
co
m
e
f
o
r
v
ar
i
o
u
s
t
h
r
es
h
o
l
d
s
.
F
u
r
t
h
er
,
t
h
e
o
p
t
i
m
i
zat
i
o
n
s
ear
c
h
co
n
v
er
g
e
n
ce b
y
M
O
A
i
s
b
et
t
e
r
co
m
p
a
r
e
d
to
o
th
e
r
a
lg
o
r
ith
m
s
c
o
n
s
id
e
r
e
d
in
t
h
is
r
e
s
e
a
r
c
h
.
F
in
a
ll
y
,
a
s
ta
tis
tic
a
l
c
o
n
f
ir
m
a
tio
n
w
it
h
W
i
l
c
ox
on
r
a
n
k t
e
s
t
i
s
t
h
e
n
pe
r
f
or
m
e
d
t
o
j
u
s
t
i
f
y
t
h
e
pe
r
f
or
m
a
n
c
e
o
f
M
O
A
c
o
m
pa
r
e
d t
o
o
t
he
r
he
ur
i
s
t
i
c
a
l
go
r
i
t
h
m
s
o
n t
he
c
ho
s
e
n
I
M
L
T
p
r
o
b
l
e
m
.
I
n
f
ut
ur
e
,
t
he
M
O
A
c
a
n
b
e
c
o
n
s
id
e
r
e
d
to
id
e
n
tif
y
th
e
f
i
ne
s
t
t
hr
e
s
ho
l
d
s
f
o
r
R
G
B
i
m
a
ge
s
a
nd
t
he
i
m
a
ge
s
c
o
r
r
up
t
e
d
w
i
t
h
no
i
s
e
.
ACK
NO
W
L
E
D
G
E
M
E
NT
S
T
h
i
s
w
or
k
w
a
s
s
u
ppor
t
e
d by
t
h
e
S
oon
c
h
unhy
a
ng
U
n
i
v
e
r
s
i
t
y
R
e
s
e
a
r
c
h
F
un
d.
R
EF
ER
EN
C
ES
[
1]
S
.
Y
as
s
i
n
e,
S
.
K
ad
r
y
,
an
d
M
.
S
i
ci
l
i
a,
"
A
f
r
a
m
e
w
o
r
k
f
o
r
l
ear
n
i
n
g
an
al
y
t
i
cs
i
n
m
o
o
d
l
e f
o
r
as
s
es
s
i
n
g
co
u
r
s
e
out
c
om
e
s
,
"
I
E
E
E
G
l
ob
al
E
ngi
ne
e
r
i
ng E
duc
at
i
o
n C
onf
.,
20
16,
p
p.
2
61
-
26
6,
doi
:
1
0.
1
10
9/
E
D
U
C
O
N
.
2
016
.
7
47
45
63.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
20
88
-
8708
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
,
V
o
l.
11
, N
o
.
6
,
D
ecem
b
er
2
0
2
1
:
5
420
-
54
29
5428
[
2]
B
.
M
ut
hu
et
a
l
.,
"
I
O
T
b
as
ed
w
e
ar
ab
l
e s
en
s
o
r
f
or
di
s
e
a
s
e
s
p
r
e
di
c
t
i
on a
nd s
y
m
pt
om
a
na
l
y
s
i
s
i
n he
a
l
t
hc
a
r
e
s
ect
o
r
,
"
P
eer
-
to
-
pe
e
r
ne
t
w
or
k
i
ng
and
ap
pl
i
c
at
i
ons
,
vo
l
.
1
3,
p
p.
2
123
-
21
34
,
J
an
.
20
20
,
do
i
:
10
.
100
7
/
s
12
08
3
-
0
19
-
008
23
-
2.
[
3]
R.
P
r
i
ya
,
V
.
R
an
g
an
at
h
an
,
S
.
K
ad
r
y
,
R
.
D
a
m
aš
e
v
i
či
u
s
,
an
d
T
.
B
l
a
žau
s
k
as
,
"
An
i
m
a
g
e en
cr
y
p
t
i
o
n
s
ch
em
e b
as
ed
o
n
bl
oc
k
s
c
r
a
m
bl
i
ng
,
m
odi
f
i
e
d z
i
g
z
a
g
t
r
a
ns
f
or
m
a
t
i
on a
nd
k
e
y
g
e
ne
r
a
t
i
on
us
i
ng
e
n
ha
nc
e
d
l
og
i
s
t
i
c
-
T
en
t
m
ap
,
"
E
nt
r
opy
,
vo
l
.
2
1,
no.
7,
p.
65
6
,
20
19
,
doi
:
1
0.
3
39
0/
e
2
10
70
65
6.
[
4]
N
.
Y
a
qoob
,
M
.
G
ul
i
s
t
a
n,
S
.
K
a
dr
y
,
a
nd H
.
A
.
W
a
ha
b,
"
C
o
m
p
le
x
in
tu
iti
o
n
is
tic
f
u
z
z
y
g
r
a
p
h
s
w
ith
a
p
p
lic
a
tio
n
i
n
cel
l
u
l
ar
n
et
w
o
r
k
p
r
o
v
i
d
er
co
m
p
an
i
es
,
"
M
a
th
e
m
a
tic
s
,
vo
l
7,
no.
1,
p.
35,
2
01
9
,
doi
:
1
0.
33
90/
m
a
t
h70
10
035
.
[
5]
M
.
A
n
b
ar
as
an
e
t
a
l
.,
"
D
e
t
e
c
t
i
on
of
f
l
ood
di
s
a
s
t
e
r
s
y
s
t
e
m
ba
s
e
d on I
oT
,
bi
g
da
t
a
a
nd c
onv
ol
ut
i
o
n
al
d
eep
n
eu
r
al
ne
t
w
or
k
,
"
C
om
put
e
r
C
om
m
un
i
c
at
i
ons
,
vol
.
15
0,
pp.
15
0
-
15
7,
J
an
.
2
020
,
d
o
i
:
1
0.
1
01
6/
j
.
c
om
c
o
m
.
2019
.
11.
02
2.
[
6]
A.
Ah
i
l
a
n
e
t
al
.
,
"
S
e
g
m
e
n
ta
tio
n
b
y
F
r
a
c
tio
n
a
l O
r
d
e
r
D
a
r
w
in
ia
n
P
a
r
tic
le
S
w
a
r
m
O
p
tim
iz
a
tio
n
B
a
s
e
d
M
u
lti
le
v
e
l
T
hr
e
s
hol
di
ng
a
nd I
m
pr
ov
e
d L
os
s
l
e
s
s
P
r
e
di
c
t
i
o
n B
a
s
e
d C
om
pr
e
s
s
i
on A
l
g
or
i
t
hm
f
or
M
e
di
c
a
l
I
m
a
g
e
s
,
"
i
n
I
E
E
E
A
cces
s
,
v
ol
.
7,
pp.
8
95
70
-
89
58
0,
201
9,
do
i
:
10.
11
09/
A
C
C
E
S
S
.
20
1
9.
2
89
16
32.
[
7]
V
.
R
a
jin
ik
a
n
t
h
,
N
. S
. M
. R
a
j
a
, a
n
d
S
.
C
.
S
at
ap
at
h
y
,
"
R
obus
t
c
ol
o
r
i
m
a
ge
m
ul
t
i
-
t
hr
e
s
hol
di
ng
us
i
ng
be
t
w
e
e
n
-
cl
as
s
v
ar
i
an
ce
an
d
cu
ck
o
o
s
ear
ch
al
g
o
r
i
t
h
m
,
" I
n
I
nf
or
m
at
i
on s
y
s
t
e
m
s
de
s
i
gn a
nd i
n
t
e
l
l
i
ge
nt
ap
pl
i
c
a
t
i
o
ns
,
Spr
i
n
ge
r
,
N
e
w
D
e
lh
i,
v
ol
.
4
33,
pp
.
3
79
-
38
6,
2
016
,
do
i
:
10
.
10
07
/
97
8
-
81
-
3
22
-
27
55
-
7_
40.
[
8]
V
.
R
a
ji
n
ik
a
n
th
,
S
.
C
.
S
at
ap
at
h
y
,
N
.
D
e
y
,
an
d
R
.
V
i
j
ay
ar
aj
an
,
"
DW
T
-
P
C
A
i
m
a
g
e
f
us
i
on t
e
c
hni
que
t
o
i
m
pr
ov
e
s
eg
m
en
t
at
i
o
n
accu
r
acy
i
n
b
r
ai
n
t
u
m
o
r
an
al
y
s
i
s
,
"
i
n
M
i
cr
o
el
ect
r
o
n
i
cs
,
el
ect
r
o
m
a
g
n
et
i
cs
a
n
d
te
le
c
o
m
m
u
n
ic
a
tio
n
s
,
Spr
i
n
ge
r
,
S
i
ng
a
por
e
,
vol
.
47
1,
pp
.
4
53
-
46
2,
20
18
,
doi
:
1
0.
10
07/
97
8
-
9
81
-
10
-
73
29
-
8_
46
.
[
9]
K
.
S
ek
ar
an
,
P
.
C
ha
n
da
na
,
N
.
M
.
K
r
i
s
hna
,
a
n
d
S
.
K
a
dr
y
,
"
D
e
e
p l
e
a
r
ni
ng
c
o
nv
ol
u
t
i
o
na
l
ne
ur
a
l
ne
t
w
or
k
(
C
N
N
)
w
ith
G
a
us
s
i
a
n m
i
x
t
ur
e
m
ode
l
f
or
pr
e
di
c
t
i
ng
p
an
cr
eat
i
c can
cer
,
"
M
ul
t
i
m
e
di
a T
o
ol
s
an
d A
p
pl
i
c
at
i
ons
,
vo
l
.
79
,
no
.
1
5,
pp.
10
23
3
-
10
24
7,
20
20
,
do
i
:
10.
1
007
/
s
11
04
2
-
01
9
-
74
19
-
5.
[
1
0]
F
.
K
u
l
a
ko
v,
S
. K
a
d
r
y
,
G
.
A
l
f
e
r
o
v
, a
n
d
A
. S
h
a
r
l
a
y
,
"
B
i
l
at
er
al
r
e
m
o
t
e co
n
t
r
o
l
o
v
er
s
p
ace
m
an
i
p
u
l
at
o
r
s
,
"
i
n
AI
P
C
o
n
f
er
en
ce P
r
o
ceed
i
ng
s
,
A
I
P
P
u
bl
i
s
hi
ng
L
L
C
,
v
ol
.
20
40,
n
o.
1,
20
18,
do
i
:
10.
10
63/
1.
50
79
21
8.
[
1
1]
Y
.
W
a
n
g
e
t a
l.
,
"
M
or
ph
ol
og
i
c
a
l
s
e
gm
e
nt
a
t
i
on a
na
l
y
s
i
s
a
nd t
e
x
t
ur
e
-
b
as
ed
s
u
p
p
o
r
t
v
ect
o
r
m
ach
i
n
es
cl
as
s
i
f
i
cat
i
o
n
o
n
m
i
c
e l
i
v
er
f
i
b
r
o
s
i
s
m
i
cr
o
s
co
p
i
c i
m
ag
es
,
"
C
u
r
r
e
n
t Bio
in
fo
r
m
a
tic
s
,
vo
l
.
1
4,
n
o.
4,
pp.
28
2
-
2
94
,
2
019
,
doi
:
1
0.
2
17
4/
1
57
48
93
61
46
66
19
0
304
12
52
21
.
[
1
2]
T.
D.
V
.
S
h
r
ee,
K
.
R
ev
an
th
,
N.
S
.
M
.
R
aj
a
,
a
nd
V.
R
a
jin
ik
a
n
th
,
"
A
h
y
br
i
d i
m
a
g
e
pr
oc
e
s
s
i
ng
a
ppr
oa
c
h t
o e
x
a
m
i
ne
a
b
n
o
r
m
a
lity
in
r
e
tin
a
l o
p
tic
d
is
c
,
"
P
r
o
ced
i
a
C
o
m
p
u
t
er
S
ci
en
ce,
vo
l
.
12
5,
pp
.
1
57
-
16
4
,
2
018
,
doi
:
1
0.
1
01
6/
j
.
pr
oc
s
.
20
17.
12
.
0
22.
[
1
3]
V
.
S
.
L
a
k
s
h
mi
,
S
.
G
.
T
e
bby
,
D
.
S
hr
i
r
a
nj
a
ni
,
a
n
d
V
.
R
a
j
i
ni
k
a
nt
h,
"
C
h
ao
t
i
c c
u
ck
o
o
s
ear
ch
an
d
K
ap
u
r
/
T
s
al
l
i
s
a
ppr
oa
c
h i
n s
e
g
m
e
nt
a
t
i
on of
T.
C
ru
zi
f
r
o
m
b
l
o
o
d
s
m
ear
i
m
ag
es
,
"
I
nt
e
r
nat
i
on
al
J
our
n
al
of
C
om
p
ut
e
r
S
ci
en
ce a
n
d
I
nf
or
m
at
i
o
n Se
c
ur
i
t
y
(
I
J
C
SI
S)
,
vol
.
1
4,
pp.
51
-
5
6,
20
16
.
[
1
4]
V
.
R
a
jin
ik
a
n
t
h
,
K
.
P
.
T
h
an
ar
aj
,
S
.
C
.
S
at
ap
at
h
y
,
S
. L
.
F
er
n
an
d
es
,
a
nd N
.
De
y
,
"
S
ha
nnon’
s
e
nt
r
opy
a
nd w
a
t
e
r
s
he
d
a
lg
o
r
ith
m
-
b
as
ed
t
e
c
hni
que
t
o i
ns
pe
c
t
i
s
c
he
m
i
c
s
t
r
ok
e
w
ound
,
"
i
n
Sm
ar
t
i
nt
e
l
l
i
ge
nt
c
om
p
ut
i
ng
an
d ap
pl
i
c
a
t
i
o
ns
,
Spr
i
n
ge
r
,
S
i
ng
a
por
e
,
vol
.
10
5,
p
p.
23
-
31
,
20
19
,
doi
:
1
0.
10
07/
97
8
-
98
1
-
13
-
19
27
-
3_
3.
[
1
5]
R.
Ro
u
f
a
y
e
l
a
nd S
.
K
a
dr
y
,
"
E
x
pr
e
s
s
i
on of
m
i
R
-
23a
b
y
a
popt
o
t
i
c
r
e
g
ul
a
t
or
s
i
n hum
a
n
c
a
nc
e
r
:
A
r
e
v
i
e
w
,
"
C
a
n
cer
bi
ol
ogy
&
t
he
r
apy
,
vo
l
.
18,
n
o.
5,
pp.
2
69
-
2
7
6,
20
17
,
do
i
:
10.
10
80/
153
84
04
7.
20
17.
13
10
34
2.
[
1
6]
V
.
R
a
ji
n
ik
a
n
th
, N
.
S
. M
. R
a
j
a
, a
n
d
K
. L
a
t
h
a
,
"
O
p
tim
a
l
m
u
ltile
v
e
l
im
a
g
e
th
r
e
s
h
o
ld
in
g
:
A
n
a
n
a
ly
s
is
w
ith
P
S
O
a
n
d
B
F
O
a
lg
o
r
ith
m
s
,
"
A
us
t
r
al
i
a
n
J
o
u
r
nal
of
B
as
i
c
a
nd A
p
pl
i
e
d Sc
i
e
nc
e
s
,
vo
l
.
8,
no
.
9
,
p
p.
44
3
-
45
4,
20
1
4.
[
1
7]
F
.
H
azzaa
an
d
S
.
K
ad
r
y
,
"
Ne
w
s
y
st
e
m
o
f
E
-
v
ot
i
ng
us
i
ng
f
i
ng
e
r
pr
i
nt
,
"
I
n
t
e
r
nat
i
on
al
J
o
ur
na
l
of
E
m
e
r
gi
ng
T
e
c
hnol
ogy
an
d A
dv
anc
e
d
E
n
gi
n
e
e
r
i
ng,
vo
l
.
2,
n
o.
10,
p
p.
35
5
-
363
,
20
12
.
[
1
8]
S.
L
.
F
er
n
an
d
es
,
U
.
J
.
T
a
n
ik
,
V
.
R
a
jin
ik
a
n
th
,
a
n
d
K
. A
. K
a
r
t
h
i
k
,
"A
r
el
i
ab
l
e f
r
a
m
e
w
o
r
k
f
o
r
accu
r
at
e b
r
ai
n
i
m
ag
e
e
x
a
m
i
na
t
i
on a
nd t
r
e
a
t
m
e
nt
pl
a
nni
ng
ba
s
e
d on e
a
r
l
y
di
a
g
nos
i
s
s
upp
or
t
f
or
c
l
i
ni
c
i
a
ns
,
"
N
e
ur
al
C
om
put
i
n
g
a
n
d
A
ppl
i
c
at
i
ons
,
v
o
l.
32
,
n
o.
20
,
p
p.
158
97
-
1
590
8,
20
20
,
do
i
:
10
.
10
07/
s
0
0
52
1
-
0
19
-
04
36
9
-
5.
[
1
9]
N.
Ot
su
,
"
A
t
hr
e
s
hol
d s
e
l
e
c
t
i
on
m
e
t
hod f
r
om
g
r
a
y
-
le
v
e
l h
is
to
g
r
a
m
s
,
"
I
E
E
E
t
ra
n
sa
c
t
i
o
n
s o
n
sy
st
e
ms,
ma
n
,
a
n
d
cyb
er
n
et
i
cs
,
vo
l
.
9,
n
o.
1,
pp.
62
-
66,
19
79
,
d
o
i
:
1
0.
1
10
9/
T
S
M
C
.
1
9
79.
43
10
07
6.
[
2
0]
V
.
R
aj
i
n
i
k
an
t
h
an
d
M
.
S
.
C
o
u
cei
r
o
,
"
O
p
tim
a
l m
u
ltile
v
e
l i
m
a
g
e
th
r
e
s
h
o
ld
s
e
le
c
tio
n
u
s
i
n
g a
no
ve
l
o
bj
e
c
t
i
ve
f
unc
t
i
on
,
" I
n
I
nf
or
m
at
i
on Sy
s
t
e
m
s
D
e
s
i
gn an
d I
n
t
e
l
l
i
ge
n
t
A
p
pl
i
c
at
i
ons
,
Spr
i
n
ge
r
,
N
ew
D
el
h
i
,
vo
l
.
34
0,
pp
.
17
7
-
18
6,
201
5
,
do
i
:
10.
10
07/
97
8
-
81
-
32
2
-
2
247
-
7_
19
.
[
2
1]
N.
S.
M
.
R
aj
a,
V
.
R
a
j
i
ni
k
a
nt
h,
a
nd K
.
L
a
t
ha
,
"
O
t
s
u ba
s
e
d opt
i
m
a
l
m
ul
t
i
l
e
v
e
l
i
m
a
g
e
t
hr
e
s
hol
di
ng
us
i
ng
f
i
r
e
f
l
y
a
lg
o
r
ith
m
,
"
M
ode
l
l
i
ng
an
d S
i
m
ul
at
i
o
n i
n E
ngi
ne
e
r
i
ng,
vo
l
.
20
14
,
2
014
,
d
oi
:
1
0.
1
15
5/
2
01
4/
79
45
74.
[
2
2]
M.
Sh
a
r
i
f
,
M
. A
. K
h
a
n
, Z
. I
q
b
a
l
,
M
.
F
.
Az
a
m
,
M
. I
.
U
.
La
l
i
,
a
nd
M
.
Y
.
J
av
ed
,
"
D
et
ect
i
o
n
an
d
cl
as
s
i
f
i
cat
i
o
n
o
f
ci
t
r
u
s
di
s
e
a
s
e
s
i
n a
g
r
i
c
ul
t
ur
e
ba
s
e
d on o
pt
i
m
i
z
e
d
w
e
i
g
ht
e
d s
e
gm
e
nt
a
t
i
o
n a
nd f
e
a
t
ur
e
s
e
l
e
c
t
i
on
,
"
C
om
put
e
r
s
an
d
e
l
e
c
t
r
oni
c
s
i
n agr
i
c
ul
t
ur
e
,
vo
l
.
150
,
pp
.
2
20
-
23
4,
J
u
l.
20
18
,
do
i
:
10.
101
6/
j
.
c
om
pa
g
.
2018
.
0
4.
0
23
.
[
2
3]
S
.
L
.
F
e
r
na
nde
s
,
V
.
R
a
j
i
ni
k
a
nt
h
,
a
nd S
.
K
a
dr
y
,
"
A
H
y
br
i
d F
r
a
m
e
w
or
k
t
o E
v
a
l
ua
t
e
B
r
e
a
s
t
A
bnor
m
a
l
i
t
y
U
s
i
ng
I
n
f
r
ar
ed
T
h
er
m
al
I
m
a
g
es
,
"
i
n
I
E
E
E
C
o
ns
um
e
r
E
l
e
c
t
r
oni
c
s
M
a
gaz
i
ne
,
vol
.
8,
no
.
5,
pp
.
31
-
36
,
S
e
p
.
20
19,
doi
:
1
0.
1
10
9/
M
C
E
.
2
01
9.
2
92
39
26
.
[
2
4]
N.
S.
M
.
R
aj
a,
K.
S
.
M
an
i
c,
an
d
V
.
R
aj
i
n
i
k
an
t
h
,
"
F
ir
e
f
ly
a
l
g
o
r
ith
m
w
ith
v
a
r
io
u
s
r
a
n
d
o
m
iz
a
tio
n
p
a
r
a
m
e
te
r
s
:
A
n
an
al
y
s
i
s
,
"
i
n
I
nt
e
r
nat
i
on
al
C
o
nf
e
r
e
nc
e
on Sw
ar
m
,
E
v
ol
ut
i
on
ar
y
,
and M
e
m
e
t
i
c
C
om
pu
t
i
n
g
,
S
pr
i
nge
r
,
C
ha
m
,
vo
l
.
82
97
,
20
13,
pp
.
1
10
-
1
21,
d
o
i:
1
0.
10
07
/
97
8
-
3
-
31
9
-
0
37
53
-
0_
11
.
[
2
5]
I.
T
.
R
oopi
n
i
,
M
.
V
as
an
t
h
i
,
V
.
R
a
j
i
n
i
k
an
t
h
,
M
.
R
ek
h
a,
an
d
M
.
S
an
g
eet
h
a,
"
S
e
gm
e
nt
a
t
i
on of
t
um
or
f
r
om
br
a
i
n M
R
I
u
s
i
n
g
f
u
zz
y
en
t
r
o
p
y
an
d
d
i
s
t
an
ce r
eg
u
l
ar
i
s
ed
l
ev
el
s
et
,
"
i
n
C
o
m
put
a
t
i
o
nal
S
i
g
nal
P
r
oc
e
s
s
i
ng
and
A
nal
y
s
i
s
,
Spr
i
n
ge
r
,
S
i
ng
a
por
e
,
vol
49
0
,
p
p.
297
-
3
04,
20
18
,
do
i
:
10
.
10
07
/
97
8
-
981
-
10
-
83
54
-
9_
27
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t
J
E
l
e
c
&
C
o
m
p
E
n
g
I
S
S
N
:
2088
-
8708
I
m
age
m
ul
t
i
-
l
evel
-
th
r
e
s
h
o
ld
in
g
w
ith
M
a
y
fly
o
p
tim
iz
a
tio
n
(
Se
i
f
e
di
ne
K
adr
y
)
5429
[
2
6]
D
.
S
h
r
ir
a
n
ja
n
i,
S
. G
. T
e
b
b
y
, S
.
C
. S
a
t
a
p
a
t
h
y
, N
. D
e
y
, a
n
d
V
. R
a
j
i
n
i
k
a
n
t
h
,
"
K
a
pur
’
s
e
nt
r
o
py
a
nd a
c
t
i
v
e
c
ont
our
-
ba
s
e
d s
e
gm
e
nt
a
t
i
on a
nd a
na
l
y
s
i
s
of
r
e
t
i
na
l
opt
i
c
di
s
c
,
"
i
n
C
om
p
ut
a
t
i
on
al
Si
g
nal
P
r
oc
e
s
s
i
ng
an
d A
na
l
y
s
i
s
,
Spr
i
nge
r
,
S
i
ng
a
por
e
,
vol
.
4
90
,
pp.
28
7
-
29
5,
20
18
,
do
i
:
1
0.
10
07
/
97
8
-
9
81
-
10
-
8
354
-
9_
26
.
[
2
7]
S.
C
.
S
at
ap
at
h
y
,
N
. S
. M
. R
a
j
a
,
V
. R
a
j
i
n
i
k
a
n
t
h
, A
. S
. A
s
h
o
u
r
, a
n
d
N
. D
e
y
,
"
M
u
lti
-
l
e
v
e
l
i
m
a
ge
t
hr
e
s
hol
di
ng
us
i
ng
O
t
s
u
an
d
ch
ao
t
i
c b
at
al
g
o
r
i
th
m
,
"
N
e
ur
al
C
om
p
ut
i
n
g a
nd
A
p
pl
i
c
at
i
o
ns
,
vo
l
.
2
9,
n
o.
12,
pp
.
12
85
-
13
07
,
20
18
,
doi
:
1
0.
1
00
7/
s
0
05
21
-
0
16
-
26
45
-
5.
[
2
8]
K
.
Ze
r
v
o
u
d
a
k
is
a
n
d
T
.
S
te
lio
s
,
"
A
m
a
y
f
l
y
o
p
tim
iz
a
tio
n
a
lg
o
r
ith
m
,
"
C
om
put
e
r
s
&
I
ndus
t
r
i
al
E
ngi
ne
e
r
i
ng
,
v
ol
.
14
5,
p.
1
06
55
9,
20
20
,
d
o
i
:
1
0.
1
01
6/
j
.
c
i
e
.
202
0.
10
65
59.
[
2
9]
T
. B
h
a
t
t
a
c
h
a
r
y
y
a
, B
. C
h
a
t
t
e
r
j
e
e
,
P
. K
. S
i
n
g
h
, J
. H
. Y
o
o
n
, Z
. W
. G
e
e
m
,
a
nd R
.
S
a
r
k
a
r
,
"
M
a
y
f
l
y
i
n H
a
r
m
ony
:
A
N
e
w
H
y
b
r
id
M
e
ta
-
H
e
u
r
is
tic
F
e
a
tu
r
e
S
e
le
c
tio
n
A
l
g
o
r
ith
m
,
"
in
I
E
E
E
A
cces
s
,
v
ol
.
8,
pp.
19
59
29
-
1
95
94
5
,
20
20,
doi
:
1
0.
1
10
9/
A
C
C
E
S
S
.
20
20
.
30
317
1
8.
[
3
0]
Z
. W
a
n
g
, A
. C
. B
o
v
i
k
, H
. R
. S
h
e
i
k
h
,
a
n
d
E.
P
.
S
im
o
n
c
e
lli,
"
I
m
a
g
e
q
u
a
lit
y
a
s
s
e
s
s
m
e
n
t: f
r
o
m
e
r
r
o
r
v
is
ib
ili
ty
to
s
tr
u
c
tu
r
a
l s
im
ila
r
it
y
,
"
in
I
E
E
E
T
r
ans
ac
t
i
ons
on
I
m
age
P
r
oc
e
s
s
i
ng
, v
o
l
.
13,
n
o.
4
,
p
p.
60
0
-
6
12,
A
pr
.
2
00
4,
doi
:
10
.
11
09
/
T
I
P
.
20
03
.
81
98
61
.
[
3
1]
A
.
H
o
r
é
a
n
d
D
.
Zio
u
,
"
I
m
a
g
e
Q
u
a
lity
M
e
tr
ic
s
: P
S
N
R
v
s
.
S
S
I
M
,
"
201
0 20
t
h I
nt
e
r
n
at
i
on
al
C
o
nf
e
r
e
nc
e
on P
at
t
e
r
n
R
e
c
ogni
t
i
o
n
,
20
10,
p
p.
23
66
-
23
6
9,
d
oi
:
1
0.
1
10
9/
I
C
P
R
.
20
10
.
5
79.
[
3
2]
D
. Z
h
a
o
e
t a
l.
,
"
C
ha
ot
i
c
r
a
ndom
s
pa
r
e
a
nt
c
ol
ony
opt
i
m
i
z
a
t
i
on
f
o
r
m
u
lti
-
t
hr
e
s
hol
d i
m
a
g
e
s
e
gm
e
nt
a
t
i
on of
2D
K
a
pur
e
nt
r
opy
,
"
K
now
l
e
d
ge
-
B
a
s
ed
S
ys
t
em
s
,
vo
l
.
21
6,
M
ar
.
20
20
,
A
r
t
.
no
.
106
51
0,
d
o
i:
10.
10
16/
j
.
k
nos
y
s
.
20
20.
10
65
10.
[
3
3]
L
.
Z
ha
ng
,
L
.
Z
ha
ng
,
X
.
M
ou,
a
n
d
D
.
Z
ha
ng
,
"
F
S
I
M
:
A
F
e
a
t
ur
e
S
i
m
i
l
a
r
i
t
y
I
nde
x
f
or
I
m
a
g
e
Q
ua
l
i
t
y
A
s
s
e
s
s
m
e
nt
,
"
i
n
I
E
E
E
T
r
ans
ac
t
i
o
ns
on I
m
age
P
r
oc
e
s
s
i
ng
,
v
ol
.
20,
no.
8,
pp.
23
78
-
23
86,
A
ug
.
201
1
,
doi
:
1
0.
1
10
9/
T
I
P
.
2
01
1.
2
10
97
30
.
[
3
4]
N.
S.
M
.
R
aj
a a
n
d
V
.
R
aj
i
n
i
k
an
t
h
,
"
B
r
ow
ni
a
n di
s
t
r
i
bu
t
i
o
n g
ui
de
d
b
a
c
t
e
r
i
a
l
f
or
a
g
i
ng
a
l
g
or
i
t
hm
f
or
c
ont
r
ol
l
e
r
de
s
i
g
n
pr
o
bl
e
m
,
"
i
n
I
C
T
an
d C
r
i
t
i
c
al
I
nf
r
as
t
r
uc
t
ur
e
:
P
r
oc
e
e
di
ngs
of
t
he
48t
h A
nn
ual
C
o
nv
e
nt
i
o
n
of
C
om
put
e
r
S
oc
i
e
t
y
o
f
I
ndi
a
-
V
o
l
I
,
Spr
i
nge
r
,
C
ham
, v
ol
.
248
,
20
14
,
p
p.
14
1
-
1
48
,
d
o
i:
10.
1
007
/
9
78
-
3
-
31
9
-
0
31
07
-
1_
17
.
[
3
5]
V
.
R
a
jin
ik
a
n
t
h
a
n
d
K
.
L
a
th
a
,
"
B
a
c
te
r
ia
l f
o
r
a
g
in
g
o
p
tim
iz
a
tio
n
a
lg
o
r
ith
m
b
a
s
e
d
P
I
D
c
o
n
tr
o
lle
r
tu
n
i
n
g
f
o
r
tim
e
de
l
a
y
e
d uns
t
a
bl
e
s
y
s
t
e
m
,
"
T
he
M
e
di
t
e
r
r
ane
an J
our
n
al
of
M
e
as
ur
e
m
e
nt
a
nd C
on
t
r
ol
,
vo
l
.
7,
n
o.
1,
pp
.
1
97
-
20
3,
201
1
.
[
3
6]
V.
R
a
j
i
n
i
k
a
nt
h
a
nd M
.
S
.
C
ouc
e
i
r
o
,
"
RG
B
h
is
to
g
r
a
m
-
b
a
s
ed
c
ol
or
i
m
a
ge
s
e
gm
e
nt
a
t
i
o
n us
i
ng
f
i
r
e
f
l
y
a
lg
o
r
ith
m
,
"
P
r
o
ced
i
a
C
o
m
p
u
t
er
S
ci
en
ce
,
vo
l
.
4
6,
pp.
14
49
-
1
45
7,
2
015
,
d
oi
:
1
0.
1
01
6/
j
.
pr
oc
s
.
20
15.
02
.
06
4.
[
3
7]
S
.
M
ir
ja
lili
,
"
M
o
th
-
f
la
m
e
o
p
ti
m
iz
a
tio
n
a
lg
o
r
ith
m
: A
n
o
v
e
l
na
t
ur
e
-
i
ns
pi
r
e
d he
ur
i
s
t
i
c
pa
r
a
di
g
m
,
"
K
now
l
e
dge
-
bas
e
d
sy
st
e
ms
,
vo
l
.
89,
pp
.
22
8
-
24
9,
No
v
.
201
5
,
do
i
:
10.
10
16/
j
.
k
nos
y
s
.
20
15.
07.
00
6.
[
3
8]
V.
J
ai
s
w
al
,
V
.
S
h
ar
m
a,
an
d
S
.
V
a
r
m
a,
"
MMFO
:
M
o
di
f
i
e
d m
ot
h f
l
a
m
e
opt
i
m
i
z
a
t
i
on a
l
g
or
i
t
hm
f
or
r
e
gi
on
ba
s
e
d R
G
B
co
l
o
r
i
m
ag
e
s
eg
m
en
t
at
i
o
n
,
"
I
n
te
r
n
a
ti
o
n
a
l J
o
u
r
n
a
l o
f
Ele
c
tr
ic
a
l
and
C
om
p
ut
e
r
E
n
gi
ne
e
r
i
n
g
(
IJ
E
CE
)
,
vo
l
.
10
,
n
o.
1,
pp.
20
88
-
8
70
8,
F
eb
.
20
20
,
do
i
:
10
.
11
59
1/
i
j
e
c
e
.
v
10i
1.
p
p1
96
-
20
1.
[
3
9]
K
.
S
ad
d
am
i
,
K
.
M
u
na
di
,
Y
.
Awa
y
,
a
n
d
F
.
Ar
n
i
a
,
"
I
m
pr
ov
e
m
e
nt
of
bi
na
r
i
z
a
t
i
on
pe
r
f
or
m
a
nc
e
us
i
ng
lo
c
a
l O
ts
u
t
hr
e
s
hol
di
ng
,
"
I
nt
e
r
nat
i
on
al
J
our
nal
of
E
l
e
c
t
r
i
c
al
a
nd C
om
pu
t
e
r
E
ngi
ne
e
r
i
ng
(
IJ
E
CE
)
,
vo
l
.
9,
n
o.
1,
pp.
26
4
-
2
72
,
201
9
,
do
i
:
10.
11
59
1/
i
j
e
c
e
.
v
9i
1
.
p
p
264
-
2
72.
[
4
0]
F
.
J.
E
s
t
r
ad
a an
d
A
.
D
.
J
e
ps
on,
"
B
e
nc
hm
a
r
k
i
ng
i
m
a
ge
s
e
gm
e
nt
a
t
i
on a
l
g
or
i
t
hm
s
,
"
I
n
te
r
n
a
tio
n
al
j
o
u
r
nal
of
c
om
p
ut
e
r
v
is
io
n
,
v
o
l.
85
,
n
o.
2,
pp
.
1
67
-
18
1,
2
00
9
,
do
i
:
10
.
10
07
/
s
11
26
3
-
0
09
-
025
1
-
z.
[
4
1]
"
B
er
k
el
e
y
S
e
g
m
en
t
at
i
o
n
D
at
as
et
:
I
m
a
g
es
"
,
200
3
.
A
c
c
e
s
s
e
d
o
n N
o
v 20
,
2
02
0.
[
O
n
l
i
n
e]
.
A
v
ai
l
ab
l
e:
h
ttp
s
:/
/w
w
w
2
.
e
e
c
s
.
b
e
r
k
e
le
y
.
e
d
u
/R
e
s
e
a
r
c
h
/P
r
o
je
c
ts
/C
S
/v
is
io
n
/
b
s
d
s
/B
S
D
S
3
0
0
/
h
tm
l/d
a
ta
s
e
t/im
a
g
e
s
.
h
tm
l
[
4
2]
"
D
at
as
et
o
f
S
t
an
d
ar
d
5
1
2
x
5
1
2
G
r
a
y
s
c
al
e
T
es
t
I
m
a
g
es
"
,
2020.
A
c
c
e
s
s
e
d on N
ov
20,
20
20
.
[
O
n
lin
e
]
.
A
v
a
ila
b
le
:
h
ttp
:
/
/
d
ecs
ai
.
u
g
r
.
es
/
cv
g
/
C
G
/
b
as
e.
h
t
m
[
4
3]
K
.
Z
er
v
o
u
d
ak
i
s
an
d
S
.
T
s
af
ar
ak
i
s
,
"
M
a
y
f
l
y
o
p
tim
iz
a
tio
n
A
lg
o
r
ith
m
-
M
at
l
ab
co
d
e
,"
M
en
d
el
ey D
a
t
a
, v
o
l.
1,
20
20
,
doi
:
1
0.
1
76
32/
5w
58s
8
hhz
2
.
1.
[
4
4]
A.
K
.
B
ha
nda
r
i
,
"
A
n
o
v
e
l b
e
ta
d
if
f
e
r
e
n
tia
l e
v
o
lu
tio
n
a
lg
o
r
ith
m
-
b
a
s
e
d
f
a
s
t
m
u
ltile
v
e
l th
r
e
s
h
o
ld
in
g
f
o
r
c
o
lo
r
im
a
g
e
s
eg
m
en
t
at
i
o
n
,
"
N
e
ur
a
l
C
om
p
ut
i
n
g an
d A
ppl
i
c
at
i
o
ns
,
vo
l
.
32
,
n
o.
9
,
pp.
45
83
-
4
61
3,
20
20
,
d
o
i:
1
0.
10
07/
s
0
05
21
-
0
18
-
377
1
-
z.
[
4
5]
M
. A
b
d
E
l
A
z
i
z
,
A
.
A
. E
w
e
e
s
,
a
n
d
A
. E
.
Ha
ssa
n
i
e
n
,
"
W
h
a
le
o
p
tim
iz
a
tio
n
a
lg
o
r
ith
m
a
n
d
m
o
th
-
f
la
m
e
o
p
tim
iz
a
tio
n
f
o
r
m
u
ltile
v
e
l th
r
e
s
h
o
ld
in
g
im
a
g
e
s
e
g
m
e
n
ta
tio
n
,
"
E
x
pe
r
t
Sy
s
t
e
m
s
w
i
t
h A
ppl
i
c
at
i
o
ns
,
vo
l
.
83
,
pp.
24
2
-
2
56,
O
ct
.
20
17
,
d
oi
:
10.
10
16
/
j
.
e
s
w
a
.
2017.
04.
02
3.
[
4
6]
M
. A
b
d
E
l
A
z
i
z
, A
. A
. E
w
e
e
s
, A
. E
. H
a
s
s
a
n
i
e
n
, M
. M
u
d
h
s
h
,
a
n
d
S
.
X
i
o
n
g
,
"
M
u
lti
-
o
b
je
c
tiv
e
w
h
a
le
o
p
tim
iz
a
tio
n
a
lg
o
r
ith
m
f
o
r
m
u
ltile
v
e
l th
r
e
s
h
o
l
d
in
g
s
e
g
m
e
n
ta
tio
n
,
"
i
n
A
dv
a
nc
e
s
i
n s
o
f
t
c
om
put
i
n
g
and
m
ac
h
i
ne
l
e
ar
ni
n
g
i
n
i
m
a
ge
p
ro
c
e
ssi
n
g
,
Spr
i
nge
r
,
C
ham
,
v
ol
.
730
,
p
p.
23
-
3
9,
20
18
,
do
i
:
10.
10
0
7/
9
78
-
3
-
31
9
-
6
37
54
-
9
_2.
[
4
7]
M
.
Ab
d
E
l
Az
i
z
,
D.
Ol
i
v
a
, A
.
E
.
E
w
ees
,
a
nd
S
.
X
i
ong
,
"
M
u
lti
-
le
v
e
l th
r
e
s
h
o
ld
i
n
g
-
b
as
ed
g
r
ey
s
cal
e i
m
a
g
e
s
e
gm
e
nt
a
t
i
on us
i
ng
m
ul
t
i
-
o
b
je
c
tiv
e
m
u
lti
-
v
er
s
e o
p
t
i
m
i
zer
,
"
E
x
pe
r
t
Sy
s
t
e
m
s
w
i
t
h A
ppl
i
c
at
i
ons
,
vo
l
.
12
5,
pp.
1
12
-
129
,
J
u
l.
20
19
,
do
i
:
10.
10
16/
j
.
e
s
w
a
.
2019.
01.
04
7.
[
4
8]
S.
J
.
M
o
us
a
v
i
r
a
d a
nd H
.
E
br
a
hi
m
pour
-
K
om
l
e
h,
"
M
ul
t
i
l
e
v
e
l
i
m
a
g
e
t
hr
e
s
hol
di
ng
us
i
ng
e
nt
r
o
py
o
f
hi
s
t
og
r
a
m
a
n
d
r
ecen
t
l
y
d
ev
el
o
p
ed
p
o
p
u
l
at
i
o
n
-
b
as
ed
m
et
ah
eu
r
i
s
t
i
c al
g
o
r
i
t
h
m
s
,
"
E
v
ol
ut
i
o
nar
y
I
n
te
llig
e
n
c
e
,
v
o
l.
1
0,
no
.
1
-
2
,
pp.
45
-
75
,
20
17
,
do
i
:
10.
10
07/
s
12
065
-
01
7
-
01
52
-
y.
Evaluation Warning : The document was created with Spire.PDF for Python.